Identifying B-vitamin requirements and phycosphere interactions for the HAB-forming dinoflagellate, Pyrodinium bahamense var. bahamense
2025-12-02
load packages:
library(googlesheets4) #scrapes google sheets
library(dplyr) # dataframe manipulation
library(ggplot2) # plotting package
library(lubridate)
library(tidyr)
library(readr)
library(scales)
library(growthrates)
library(patchwork)
library(grid)
library(qiime2R)
library(phyloseq)
library(viridis)
library(magrittr)
library(vegan)
library(stringr)
library(ape)
library(RColorBrewer)Create a color pallete
pp_palette <- c(
"#7D8F69", "cornflowerblue", "#F5C45E", "plum",
"hotpink4", "lightsalmon", "steelblue", "indianred", "springgreen4",
"#102E50", "maroon", "lightgreen", "lightblue", "gold",
"#FF0099", "#A26769", "blue", "red", "lightpink",
"#71816D", "#FFE6E6", "#FF66CC", "#A3B18A", "salmon", "magenta","orchid","deepskyblue", "tomato",
"cyan", "hotpink", "yellow", "orange", "navy", "darkred",
"#C0A98E", "#7A9E9F", "#A1869E", "#BFD8B8", "#F7DD72","green", "darkgreen","purple", "coral", "#FFA500",
"#FFE5AD",
"#CDB4DB", "pink","dodgerblue1","green2", "brown4", "darkviolet", "azure2", "aquamarine"
)
scales::show_col(pp_palette)
1 Characterization of bacterial communities in P. bahamense cultures
1.1 Import sequence data from QIIME 2:
Amplicons were processed on a QIIME 2 pipeline, which includes DADA denoising. Seqeunces were classified with Native Bayers Classifier
1.2 Import and Format Sample Metadata:
meta <- read.csv("SampleMetaData.csv")
meta$SampleID <- paste0(meta$SampleID, "_6283")
row.names(meta)<- meta$SampleID
meta$TreatRep <- paste(meta$PCRDate_Treat, meta$SampleInfo_Rep, sep = "_")
#keep only the P. bahamense project
pyro_only <- meta[grepl("^Pyro", meta$Project, ignore.case = TRUE), ]
META <- sample_data(pyro_only)1.3 Import and Format Taxonomy Data:
taxonomy <- read.csv("taxonomy.csv", stringsAsFactors = FALSE)
names(taxonomy) <- c("row", "tax", "Confidence")
row.names(taxonomy) <-taxonomy[[1]] #move the feature ID column to become the row names
taxonomy <- taxonomy[,(-1)] #delete the feature ID column
taxonomy <- separate(taxonomy, tax, c("Domain","Phylum", "Class", "Order", "Family", "Genus", "Species", "D7", "D8", "D9", "D10", "D11", "D12", "D13", "D14"), sep = ";", fill = "right")
taxonomy <- taxonomy[,c(1:7)] #keep first seven levels
taxonomy$D0 <- with(taxonomy, ifelse(Order == " o__Chloroplast", "Chloroplast", "Bacteria")) #create column for ordering
# Reorder columns and turn into phyloseq tax_table
col_order<- c("D0", "Domain","Phylum", "Class", "Order", "Family", "Genus", "Species")
taxonomy<- taxonomy[, col_order]
taxmat <- as.matrix(taxonomy)
TAX = tax_table(taxmat)1.4 Build a phyloseq object:
1.5 Domain-Level Relative Abundance Bar Plot:
#Keep only Bacteria + Chloroplast
ps<- subset_taxa(ps, D0 %in% c("Bacteria", "Chloroplast") )
#Convert counts to relative abundance (%)
psra<- transform_sample_counts(ps, function(OTU) 100* OTU/sum(OTU))
#Group OTU by DO level
glomD1<- tax_glom(psra, 'D0')
#plot RA
taxabarplot <- plot_bar(glomD1, x = "SampleID", fill = "D0") +
scale_y_continuous(expand = c(0, 0)) +
theme_classic() +
theme(
legend.title = element_blank(),
text = element_text(size = 14),
axis.text.x = element_text(angle = 90)
) +
ylab("Relative Abundance (%)") +
xlab("") +
scale_fill_manual(values = c("lightgrey", "#7BB03B")) +
facet_grid(~Project, scales = "free_x")
taxabarplot
1.6 Remove chloroplast sequences from data:
1.7 Calculate observed richness:
plugin <- ps_nochloro %>%
estimate_richness(measures = "Observed") %$% Observed
#Pull project metadata field from phyloseq object
Project <- ps_nochloro %>% sample_data %$% Project
#Combine richness + project into one dataframe
richness<- data.frame(plugin, Project)
names(richness) <- c("richness", "Project")
richness %>%group_by(Project) %>% summarize(mean = mean(richness), min = min(richness), max = max(richness))1.8 Relative abundance and sums the counts of all OTUs that belong to the same Order:
1.9 Bar plot at the order level:
# subset by pyro samples
pyro <- subset_samples(ps_nochloro_RA_glomO, Project == "Pyro")
# Remove low abundance orders
pyro = filter_taxa(pyro, function(x) sum(x) > 1, TRUE)
taxabarplot <- plot_bar(pyro, x= "SampleID", fill = "Order") + scale_y_continuous(expand = c(0, 0)) + ggtitle("") + theme(legend.title=element_blank()) + geom_bar(aes(fill=Order), stat="identity", position="stack", width =0.9) +theme_classic() + theme(text = element_text(size=14))+theme(axis.text.x = element_text(angle = 90)) + xlab("Sample") + ylab("Relative Abundance (%)") + theme(text = element_text(size=14)) + scale_fill_manual(values= pp_palette)+ xlab("")
taxabarplot
1.10 Bar plot at family level:
ps_nochloro_RA<- transform_sample_counts(ps_nochloro, function(OTU) 100* OTU/sum(OTU))
#Group OTU by family and sum abundances
ps_nochloro_RA_glomF<- tax_glom(ps_nochloro_RA, 'Family')
#Keep only the Pyro samples
pyro <- subset_samples(ps_nochloro_RA_glomF, Project == "Pyro")
#Remove low abundance orders
pyro = filter_taxa(pyro, function(x) sum(x) > 1, TRUE)
taxabarplot<-plot_bar(pyro, x= "SampleID", fill = "Family") + scale_y_continuous(expand = c(0, 0)) + ggtitle("") + theme(legend.title=element_blank()) + geom_bar(aes(fill=Family), stat="identity", position="stack", width =0.9) +theme_classic() + theme(text = element_text(size=14))+theme(axis.text.x = element_text(angle = 90)) + xlab("Sample") + ylab("Relative Abundance (%)") + theme(text = element_text(size=14)) + scale_fill_manual(values= pp_palette)+ xlab("")
taxabarplot
1.11 Filter Samples and Generate Family-Level Bar Plot:
#Filter SampleID
pyroE <- subset_samples(pyro, !(SampleID %in% c("L14_6283", "L15_6283", "L17_6283", "L18_6283")))
#Remove low abundance orders
pyroE = filter_taxa(pyroE, function(x) sum(x) > 1, TRUE)
taxabarplot<-plot_bar(pyroE, x= "SampleID", fill = "Family") + scale_y_continuous(expand = c(0, 0)) + ggtitle("") + theme(legend.title=element_blank()) + geom_bar(aes(fill=Family), stat="identity", position="stack", width =0.9) +theme_classic() + theme(text = element_text(size=14))+theme(axis.text.x = element_text(angle = 90)) + xlab("Sample") + ylab("Relative Abundance (%)") + theme(text = element_text(size=14)) + scale_fill_manual(values= pp_palette)+ xlab("")
taxabarplot
1.12 Determine Unique OTU’s and abundances in P. bahamense data:
## [1] "12a986d7ed4f3be699daa0feaba41d8a" "07ae9a3ebf44d4a821d001debe7d9b0c"
## [3] "e9034a72595933ebd5c5df08eb4a08fc"- 12a986d7ed4f3be699daa0feaba41d8a
o__Rhizobiales f__Hyphomicrobiaceae g__Filomicrobium s__uncultured_Alphaproteobacteria
12a986d7ed4f3be699daa0feaba41d8a GCAGCAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCCATGCCGCGTGAGTGACGAAGGCCTTAGGGTTGTAAAACTCTTTTGGCGGGGACGATAATGACGGTACCCGCAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGTTCGGAATCACTGGGCGTAAAGCGCACGTAGGCGGACTGGTCAGTTGGGGGTGAAATCCCAGGGCTCAACCCTGGAACTGCCTCCAATACTGCCAGTCTTGAGTCCGAGAGAGGTGAGTGGAATTCCTAGTGTAGAGGTGAAATTCGTAGATATTAGGAAGAACACCAGTGGCGAAGGCGGCTCACTGGCTCGGTACTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
uncultured sanger sequence from Mangrove Sediment in Hong Kong - GenBank: MH091208.1, MH091199.1 (2 best hits, 100% ) Diversity and dynamics of microbial community structure in different mangrove, marine and freshwater sediments during anaerobic debromination of PBDEs
Diversity of salt marsh prokaryotes, M.A. Moran, uncultured Hyphomicrobiaceae bacterium - lon=81.2797W, lat=31.3884N; sediment 14-16cm collected on Feb 01, 2002, Sapelo Island Microbial Observatory Dean Creek Marsh sampling site” (3rd best hit 99%)
pE_m %>% filter(OTU == "12a986d7ed4f3be699daa0feaba41d8a") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))- e0169d54945a8a584037e96365ec7bb3
o__Rhizobiales f__Hyphomicrobiaceae g__Filomicrobium s__uncultured_Alphaproteobacteria
e0169d54945a8a584037e96365ec7bb3 GCTGCAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCCATGCCGCGTGAGTGACGAAGGCCTTAGGGTTGTAAAACTCTTTTGGCGGGGACGATAATGACGGTACCCGCAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGTTCGGAATCACTGGGCGTAAAGCGCACGTAGGCGGACTGGTCAGTTGGGGGTGAAATCCCAGGGCTCAACCCTGGAACTGCCTCCAATACTGCCAGTCTTGAGTCCGAGAGAGGTGAGTGGAATTCCTAGTGTAGAGGTGAAATTCGTAGATATTAGGAAGAACACCAGTGGCGAAGGCGGCTCACTGGCTCGGTACTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
MH091208.1 uncultured bacteria - Diversity and dynamics of microbial community structure in different mangrove, marine and freshwater sediments during anaerobic debromination of PBDEs - isolation_source=“mangrove sediments” in Hong Kong, 99%
pE_m %>% filter(OTU == "e0169d54945a8a584037e96365ec7bb3") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))- aa0c627da86b1ddbba980c9dfd6ac380
o__Rhizobiales f__Hyphomicrobiaceae g__Filomicrobium s__uncultured_Alphaproteobacteria
aa0c627da86b1ddbba980c9dfd6ac380 GCAGCAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCCATGCCGCGTGAGTGACGAAGGCCTTAGGGTTGTAAAACTCTTTTGGCGGGGACGATAATGACGGTACCCGCAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGTTCGGAATCACTGGGCGTAAAGCGCACGTAGGCGGACTGGTCAGTTGGGGGTGAAATCCCAGGGCTCAACCCTGGAACTGCCTCCAATACTGCCAGTCTTGAGTCCGAGAGAGGTGAGTGGAATTCCTAGTGTAGAGGTGAAATTCGTAGATATTAGGAAGAACACCAGTGGCGAAGGCGACTCACTGGCTCGGTACTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
MH091208.1 uncultured bacteria - Diversity and dynamics of microbial community structure in different mangrove, marine and freshwater sediments during anaerobic debromination of PBDEs - Hong Kong isolation_source=“mangrove sediments” , 99%
pE_m %>% filter(OTU == "aa0c627da86b1ddbba980c9dfd6ac380") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))- f6a0f1f2d2a083164d1fb145b000b96b
o__Rhizobiales f__Hyphomicrobiaceae g__Filomicrobium s__uncultured_Alphaproteobacteria
f6a0f1f2d2a083164d1fb145b000b96b GCAGCAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCCATGCCGCGTGAGTGACGAAGGCCTTAGGGTTGTAAAACTCTTTTGGTGGGGACGATAATGACGGTACCCGCAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGTTCGGAATCACTGGGCGTAAAGCGCACGTAGGCGGACTGGTCAGTTGGGGGTGAAATCCCAGGGCTCAACCCTGGAACTGCCTCCAATACTGCCAGTCTTGAGTCCGAGAGAGGTGAGTGGAATTCCTAGTGTAGAGGTGAAATTCGTAGATATTAGGAAGAACACCAGTGGCGAAGGCGGCTCACTGGCTCGGTACTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
MH091208.1 Uncultured bacteria Diversity and dynamics of microbial community structure in different mangrove, marine and freshwater sediments during anaerobic debromination of PBDEs - isolation_source=“mangrove sediments” - 99%
pE_m %>% filter(OTU == "f6a0f1f2d2a083164d1fb145b000b96b") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))- ff08ec2d8a67e142bc459567275bb888
o__Rhizobiales f__Hyphomicrobiaceae g__Filomicrobium s__uncultured_Alphaproteobacteria
ff08ec2d8a67e142bc459567275bb888 GCAGCAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCCATGCCGCGTGAGTGACGAAGGCCTTAGGGTTGTAAAACTCTTTTGGCGGGGACGATAATGACGGTACCCGCAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGGGCTAGCGTTGTTCGGAATCACTGGGCGTAAAGCGCACGTAGGCGGACTGGTCAGTTGGGGGTGAAATCCCAGGGCTCAACCCTGGAACTGCCTCCAATACTGCCAGTCTTGAGTCCGAGAGAGGTGACTGGAATTCCTAGTGTAGAGGTGAAATTCGTAGATATTAGGAAGAACACCAGTGGCGAAGGCGGCTCACTGGCTCGGTACTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
MH091208.1 uncultured bacteria Diversity and dynamics of microbial community structure in different mangrove, marine and freshwater sediments during anaerobic debromination of PBDEs Hong Kong isolation_source=“mangrove sediments” 99%
EU488009.1 uncultured bacteria Characterization of the lucinid bivalve-bacteria symbiotic system: the significance of the geochemical habitat on bacterial symbiont diversity and phylogeny isolation_source=“siliciclastic sedment from Thalassia sea grass bed” 99%
other marine sediments, estuary sediments
pE_m %>% filter(OTU == "ff08ec2d8a67e142bc459567275bb888") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))- 07ae9a3ebf44d4a821d001debe7d9b0c
c__Acidimicrobiia o__Microtrichales f__uncultured g__uncultured NA
07ae9a3ebf44d4a821d001debe7d9b0c GCAGCAGTGGGGAATCTTGCGCAATGGGCGAAAGCCTGACGCAGCGACGCCGCGTGCGGGAAGACGGCCTTCGGGTTGTAAACCGCTTTCAGCAGGGACGAAATTGACGGTACCTGCAGAAGAAGCTCCGGCCAACTACGTGCCAGCAGCCGCGGTAAGACGTAGGGGGCGAGCGTTGTCCGGAATCATTGGGCGTAAAGGGCTCCTAGGTGGTTCAGTAAGTCGACTGTGAAAATCCAAGGCTCAACCTTGGGACGCCAGTCGATACTGCTGTGACTCGAGTTCGGTAGAGGAGTGTGGAATTCCTGGTGTAGCGGTGAAATGCGCAGATATCAGGAGGAACACCAACGGCGAAGGCAGCACTCTGGGCCGATACTGACACTGAAGAGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
KM840989.1 uncultured bacteria - Microbial diversity of indigenous bacteria in a 129I contaminated groundwater plume at the Hanford Site, Washington - Sanger, isolation_source=“iodine (I-129) contaminated groundwater” 100%
HF558551.1 uncultured bacteria - Iron- and Sulphur- cycling bacteria mobilize copper in a multiple extreme mine tailings in the Atacama Desert, Chile - isolation_source=“tailing material” - 100%
JQ427269.1 uncultured bacteria - Bacterial diversity in an alkaline saline soil spiked with anthracene, samger sequenced, isolation_source=“soil”, 100%
JQ665390.1 uncultured actinobacterium, Diversity of unculturable Actinomycetes in coastal wetlands of the Yellow River estuary, isolation_source=“soil sample from coastal wetlands”, 99%
pE_m %>% filter(OTU == "07ae9a3ebf44d4a821d001debe7d9b0c") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))- 9b8accad19c30d7a69a6290553111166
c__Acidimicrobiia o__Microtrichales f__uncultured g__uncultured NA
9b8accad19c30d7a69a6290553111166 GCTGCAGTGGGGAATCTTGCGCAATGGGCGAAAGCCTGACGCAGCGACGCCGCGTGCGGGAAGACGGCCTTCGGGTTGTAAACCGCTTTCAGCAGGGACGAAATTGACGGTACCTGCAGAAGAAGCTCCGGCCAACTACGTGCCAGCAGCCGCGGTAAGACGTAGGGGGCGAGCGTTGTCCGGAATCATTGGGCGTAAAGGGCTCCTAGGTGGTTCAGTAAGTCGACTGTGAAAATCCAAGGCTCAACCTTGGGACGCCAGTCGATACTGCTGTGACTCGAGTTCGGTAGAGGAGTGTGGAATTCCTGGTGTAGCGGTGAAATGCGCAGATATCAGGAGGAACACCAACGGCGAAGGCAGCACTCTGGGCCGATACTGACACTGAAGAGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
KM840989.1 uncultured bacteria - Microbial diversity of indigenous bacteria in a 129I contaminated groundwater plume at the Hanford Site, Washington - isolation_source=“iodine (I-129) contaminated groundwater” - 99%
pE_m %>% filter(OTU == "9b8accad19c30d7a69a6290553111166") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))- e9034a72595933ebd5c5df08eb4a08fc
o__Oceanospirillales f__Alcanivoracaceae1 g__Alcanivorax s__Alcanivorax_venustensis
e9034a72595933ebd5c5df08eb4a08fc GCAGCAGTGGGGAATCTTGGACAATGGGGGCAACCCTGATCCAGCCATGCCGCGTGTGTGAAGAAGGCCTTCGGGTTGTAAAGCACTTTCAGCAGGGAGGAAGGCTTACCCCTAATACGGGTGAGTACTTGACGTTACCTGCAGAAGAAGCACCGGCTAATTTCGTGCCAGCAGCCGCGGTAATACGAAAGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCGCGTAGGCGGTGTGTTAAGTCGGATGTGAAAGCCCAGGGCTCAACCTTGGAATTGCATCCGATACTGGCACGCTAGAGTGCAGTAGAGGGAGGTGGAATTTCCGGTGTAGCGGTGAAATGCGTAGAGATCGGAAGGAACACCAGTGGCGAAGGCGGCCTCCTGGACTGACACTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
PP516259.1 organism=“Alloalcanivorax venustensis” Exploration and conservation of bacterial community from the Arabian Sea seamount isolation_source=“Arabian Sea water”, 100%
pE_m %>% filter(OTU == "e9034a72595933ebd5c5df08eb4a08fc") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))- c02844d75d512c1510de7334b6687731
o__Rhizobiales f__Hyphomicrobiaceae g__uncultured s__uncultured_bacterium
c02844d75d512c1510de7334b6687731 GCAGCAGTGGGGAATATTGGACAATGGGCGAAAGCCTGATCCAGCCATGCCGCGTGAGTGACGAAGGCCTTAGGGTTGTAAAGCTCTTTTGGCGGGGAAGATAATGACGGTACCCGCAGAATAAGCTCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACGAAGGGAGCTAGCGTTGTTCGGAATCACTGGGCGTAAAGCGCACGTAGGCGGATTTGTTAGTCAGGGGTGAAATCCCGGGGCTCAACCCCGGAACTGCCTTTGATACTGCAAATCTCGAGTCCGAGAGAGGTGGGTGGAATTCCTAGTGTAGAGGTGAAATTCGTAGATATTAGGAAGAACACCGGTGGCGAAGGCGGCCCACTGGCTCGGTACTGACGCTGAGGTGCGAAAGCGTGGGGAGCAAACAGGATTAGATAC
OK235762.1 Uncultured Pedomicrobium sp. isolation_source=“date palm rhizosphere soil” Saudi Arabia - 99%
HQ697761.1 Uncultured bacteria isolation_source=“hydrocarbon contaminated saline-alkali soil” China - 99%
99% to several other contaminated soil sites (real and experimental)
KP098952.1 Uncultured bacteria - The microbiome of methanol-utilizing denitrification systems contains new bacterial groups - isolation_source=“denitrification bioreactor” - 99%
pE_m %>% filter(OTU == "c02844d75d512c1510de7334b6687731") %>% summarize(mean = mean(Abundance), min = min(Abundance), max = max(Abundance))1.13 Plot shannon diversity and relative abundance at order level:
# -------------------------------
# Subset Pyro samples and filter
# -------------------------------
pyro <- subset_samples(ps_nochloro_RA_glomO, Project == "Pyro")
pyro <- subset_samples(pyro, Treatment != "AB")
pyro <- filter_taxa(pyro, function(x) sum(x) > 1, TRUE)
pyro <- subset_samples(pyro,
Treatment %in% c("1003", "Replete") &
TreatRep %in% c("1003_1", "1_2"))
# Set factor levels for Notes_Filter
pyro@sam_data$Notes_Filter <- factor(as.character(pyro@sam_data$Notes_Filter),
levels = c("10", "2"))
# Clean Order names in tax_table
tax_table(pyro)[, "Order"] <- gsub(" o__", "", tax_table(pyro)[, "Order"])
# -------------------------------
# Plot relative abundance barplot
# -------------------------------
barplot <- plot_bar(pyro, x = "TreatRep", fill = "Order") +
geom_bar(aes(fill = Order), stat = "identity", position = "stack", width = 1) +
scale_y_continuous(expand = c(0, 0)) +
scale_fill_manual(values = pp_palette) +
facet_grid(Notes_Filter ~ Treatment, scales = "free") +
scale_x_discrete(
expand = c(0.5, 0.05),
labels = c(
"1003_1" = "IRL",
"1_2" = "OTB"
)
) +
theme_classic() +
theme(
legend.title = element_blank(),
legend.position = "right",
legend.justification = "center",
legend.box = "horizontal",
axis.text.x = element_text(angle = 0, vjust = 0.5, size = 16, color = "black"),
axis.text.y = element_text(size = 20, color = "black"),
axis.title.x = element_text(size = 20, color = "black"),
axis.title.y = element_text(size = 20, color = "black"),
axis.ticks.x = element_line(color = "black", linewidth = 0.9),
axis.ticks.y = element_line(color = "black", linewidth = 0.9),
strip.text.x = element_blank(), # remove top facet labels
strip.text.y = element_text(size = 20, color = "black"), # row facets
text = element_text(size = 20, color = "black")
) +
ylab(expression("Relative Abundance (%)")) +
xlab("")
barplot
# -------------------------------
# Calculate Shannon index
# -------------------------------
pyro <- subset_samples(ps_nochloro_RA_glomO, Project == "Pyro")
pyro <- subset_samples(pyro, TreatRep %in% c("1003_1", "1_2", "1_1", "1_3"))
pyro <- subset_samples(pyro, Treatment != "AB")
shannon_df <- estimate_richness(pyro, measures = "Shannon")
# Extract metadata and combine with diversity
sample_meta <- as(sample_data(pyro), "data.frame")
shannon_df <- cbind(shannon_df, sample_meta)
# -------------------------------
# Plot Shannon diversity
# -------------------------------
shannon_plot <- ggplot(shannon_df, aes(x = Treatment, y = Shannon, color = Notes_Filter)) +
# Color mapping
scale_color_manual(values = c("10" = "navy", "2" = "maroon")) +
# Points
geom_point(size = 5) +
# Custom x-axis labels
scale_x_discrete(labels = c(
"1003" = "IRL",
"Replete" = "OTB"
)) +
# Classic theme
theme_classic() +
# Axis labels
ylab("Shannon Diversity Index") +
xlab("") +
labs(color = "Filter Size") +
# Theme adjustments
theme(
legend.title = element_text(size = 20),
axis.text.x = element_text(angle = 0, vjust = 0.5, size = 16, color = "black"),
axis.text.y = element_text(size = 20, color = "black"),
axis.title.x = element_text(size = 20, color = "black"),
axis.title.y = element_text(size = 20, color = "black"),
axis.ticks.x = element_line(color = "black", linewidth = 0.9),
axis.ticks.y = element_line(color = "black", linewidth = 0.9),
strip.text.x = element_blank(), # remove top facet labels
strip.text.y = element_text(size = 20, color = "black"), # row facets
text = element_text(size = 20, color = "black"))
# Print Shannon plot
shannon_plot
1.14 Figure 2
Fig. 2 Alpha diversity and composition of bacterial communities in P. bahamense cultures. A Shannon diversity index for particle-associated (>10.0-µm size fraction; blue) and free-living (0.2–10.0-µm size fraction; red) bacterial communities in cultures of FWC1003 and FWC1006 P. bahamense isolated from Indian River Lagoon and Old Tampa Bay, respectively. B Relative abundance of bacterial orders in size-fractioned communities in P. bahamense cultures.
2 Evaluating the B-vitamin requirements of P. bahamense
Chlorophyll-a fluorescence (RFU) data from P. bahamense cultures were used to visualize growth patterns, quantify maximum biomass, and estimate specific growth rates under media conditions deficient in media lacking one targeted B vitamin. Specific growth rates were computed using the “alleasy_linear” function of the “growthrates” package in R Studio (Hall et al., 2014; RStudio Team, 2020).
2.1 Import the Dataset:
2.2 Visulaize growth:
LD_OTB_all_growth <- ggplot(
fluor %>%
filter(
Treatment %in% c("-Cobalamin", "Replete", "-Thiamine", "-Biotin"),
Round %in% c(1, 2,3,4)),
aes(x = Time, y = `Chlorophyll-A (RFU)`, color = Treatment)) +
geom_point(size = 3) + geom_smooth(se=FALSE)+
# Manual color scale
scale_color_manual(
values = c(
"-Cobalamin" = "#102E50",
"Replete" = "#F5C45E",
"-Thiamine" = "deeppink4",
"-Biotin" = "#667028"
),
labels = c(
"-Cobalamin" = expression(" —B"[12]),
"-Thiamine" = expression("—thiamine"),
"-Biotin" = expression("—biotin"),
"Replete" = "replete"),
name = "Treatment") +
labs(
x = "Time (days)",
y = expression(paste("Chlorophyll-", italic("a"), " fluorescence (RFU)"))
) +
facet_grid(~Round) +
theme_minimal() +
theme(
panel.border = element_rect(fill = NA, color = "black"),
panel.grid = element_blank(),
axis.title = element_text(size = 20),
axis.text = element_text(size = 20, color = "black"),
strip.text = element_text(size = 20, color = "black"),
legend.title = element_text(size = 18, color = "black"),
legend.text = element_text(size = 18, color = "black"),
legend.position = "bottom",
axis.ticks = element_line(color = "black", linewidth = 0.5),
axis.ticks.length = unit(4, "pt")
)
LD_OTB_all_growth
2.3 Calculate maximum RFU values for each round and medium treatment
max_bio <- fluor %>%
filter(Round %in% c(1, 2, 3, 4)) %>%
filter(Treatment %in% c("-Cobalamin", "Replete", "-Biotin", "-Thiamine")) %>%
group_by(Treatment, Round, Rep) %>%
summarise(Max_Chl = max(`Chlorophyll-A (RFU)`, na.rm = TRUE), .groups = "drop")
print(max_bio)## # A tibble: 48 × 4
## Treatment Round Rep Max_Chl
## <chr> <dbl> <dbl> <dbl>
## 1 -Biotin 1 1 2911.
## 2 -Biotin 1 2 3220.
## 3 -Biotin 1 3 6239.
## 4 -Biotin 2 1 6771.
## 5 -Biotin 2 2 4230.
## 6 -Biotin 2 3 3492.
## 7 -Biotin 3 1 4288.
## 8 -Biotin 3 2 3237.
## 9 -Biotin 3 3 1767.
## 10 -Biotin 4 1 2049.
## # ℹ 38 more rows2.3.1 Run an ANOVA on maximum RFU values and Tukey HSD post-hoc test for treatment
## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Max_Chl ~ Treatment, data = max_bio)
##
## $Treatment
## diff lwr upr p adj
## -Cobalamin--Biotin -3365.871 -4602.055 -2129.687 0.0000000
## -Thiamine--Biotin 1961.220 725.036 3197.404 0.0006407
## Replete--Biotin -757.515 -1993.699 478.669 0.3694455
## -Thiamine--Cobalamin 5327.091 4090.907 6563.275 0.0000000
## Replete--Cobalamin 2608.356 1372.172 3844.540 0.0000068
## Replete--Thiamine -2718.735 -3954.919 -1482.551 0.0000030## Df Sum Sq Mean Sq F value Pr(>F)
## Treatment 3 174966709 58322236 45.35 1.63e-13 ***
## Residuals 44 56590796 1286154
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 12.3.2 Prepare data for plotting maximum RFU values:
# Calculate mean and SD
summary_stats <- max_bio %>%
group_by(Treatment, Round) %>%
summarise(
mean_chl = mean(Max_Chl, na.rm = TRUE),
sd_chl = sd(Max_Chl, na.rm = TRUE),
.groups = "drop"
)
# Make Treatment a factor in both datasets
max_bio$Treatment <- factor(max_bio$Treatment,
levels = c( "-Biotin", "-Cobalamin","-Thiamine", "Replete"))
summary_stats$Treatment <- factor(summary_stats$Treatment,
levels = c("-Cobalamin", "-Thiamine", "-Biotin", "Replete"))
max_bio$Treatment <- factor(
max_bio$Treatment,
levels = c("-Cobalamin", "Replete", "-Thiamine", "-Biotin")
)
summary_stats$Treatment <- factor(
summary_stats$Treatment,
levels = c("-Cobalamin", "Replete", "-Thiamine", "-Biotin"))2.3.3 Plot maximum RFU values:
max_bio_LD <- ggplot(max_bio, aes(x = Treatment, y = Max_Chl, color = Treatment)) +
# Raw points (per round)
geom_point(size = 3, shape = 16, position = position_jitter(width = 0.1)) +
# Error bars: mean ± SD across replicates
geom_errorbar(
data = summary_stats,
aes(
x = Treatment,
y = mean_chl,
ymin = mean_chl - sd_chl,
ymax = mean_chl + sd_chl,
color = Treatment
),
width = 0.3,
position = position_dodge(width = 0.6),
inherit.aes = FALSE
) +
# Open circles for replicate-averaged means
geom_point(
data = summary_stats,
aes(x = Treatment, y = mean_chl, color = Treatment),
shape = 21,
size = 5,
stroke = 1.2,
position = position_dodge(width = 0.6),
inherit.aes = FALSE
) +
# Labels
labs(
x = "Media Treatment",
y = expression(paste("Maximum RFU"))
) +
# Manual colors
scale_color_manual(
values = c(
"-Cobalamin" = "#102E50",
"-Thiamine" = "deeppink4",
"-Biotin" = "#667028",
"Replete" = "#F5C45E"
),
labels = c(
"-Cobalamin" = expression("—B"[12]),
"-Thiamine" = expression("—Thiamine"),
"-Biotin" = expression("—Biotin"),
"Replete" = "Replete"
),
name = "Treatment",
guide = guide_legend(
override.aes = list(
shape = 21,
fill = "white",
stroke = 1.2,
size = 2
)
)
) +
# X-axis label formatting
scale_x_discrete(
labels = c(
"-Biotin" = expression("\u2013Biotin"),
"-Cobalamin" = expression("\u2013B"[12]),
"-Thiamine" = expression("\u2013Thiamine"),
"Replete" = "Replete"
)
) +
scale_y_continuous(labels = comma) +
# Theme
theme_minimal() +
theme(
panel.border = element_rect(fill = NA, color = "black"),
panel.grid = element_blank(),
axis.text.y = element_text(size = 20, color = "black"),
axis.text.x = element_text(angle = 45, hjust = 1, size = 16, color = "black"),
axis.title.y = element_text(size = 22),
axis.title.x = element_text(size = 20, margin = margin(t = 10)),
strip.text.x = element_text(size = 18),
legend.title = element_text( size = 20, color = "black"),
legend.text = element_text( size = 16, color = "black"),
legend.position = "none",
axis.ticks = element_line(color = "black", linewidth = 0.5),
axis.ticks.length = unit(4, "pt")
) +
guides(color = "none", fill = "none", shape = "none")+
coord_cartesian(ylim = c(0, 10000))
# Print plot
max_bio_LD
2.4 Specific growth rates using all_easylinear
Specific Growth Rates Biotin trial 1 using all_easylinear:
bio1 <- fluor %>%
filter(Treatment == "-Biotin", Round %in% c(1) )%>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
bio_lin1 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, bio1, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(bio_lin1, log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(bio_lin1)## $`1:-Biotin:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.1426 -0.3358 0.3733 0.3348 0.1439 -0.2512 -0.1225
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.66156 0.21185 17.284 1.19e-05 ***
## x 0.20749 0.02938 7.063 0.00088 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3109 on 5 degrees of freedom
## Multiple R-squared: 0.9089, Adjusted R-squared: 0.8907
## F-statistic: 49.88 on 1 and 5 DF, p-value: 0.0008799
##
##
## $`1:-Biotin:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.2611 0.1266 0.1952 0.1704 -0.1053 -0.1483 0.0224
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.77140 0.13104 28.78 9.49e-07 ***
## x 0.21617 0.01817 11.90 7.40e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1923 on 5 degrees of freedom
## Multiple R-squared: 0.9659, Adjusted R-squared: 0.959
## F-statistic: 141.5 on 1 and 5 DF, p-value: 7.396e-05
##
##
## $`1:-Biotin:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.018655 -0.066394 0.061972 0.130916 -0.079399 -0.037937 0.009497
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.479439 0.142763 17.37 1.16e-05 ***
## x 0.257843 0.007742 33.30 4.59e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.08194 on 5 degrees of freedom
## Multiple R-squared: 0.9955, Adjusted R-squared: 0.9946
## F-statistic: 1109 on 1 and 5 DF, p-value: 4.589e-07## y0 y0_lm mumax lag
## 1:-Biotin:1 33.75 38.92188 0.2074863 -0.6871588
## 1:-Biotin:2 33.46 43.44102 0.2161749 -1.2076036
## 1:-Biotin:3 49.69 11.93456 0.2578427 5.5319204## 1:-Biotin:1.r2 1:-Biotin:2.r2 1:-Biotin:3.r2
## 0.9088927 0.9658730 0.9955119Specific Growth Rates Biotin trial 2 using all_easylinear
bio2 <- fluor %>%
filter(Treatment == "-Biotin", Round %in% c(2) )%>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
bio_lin2 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, bio2, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(bio_lin2, log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(bio_lin2)## $`2:-Biotin:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -1.29805 0.90387 0.62634 0.53091 -0.40115 -0.02684 -0.33508
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.63417 1.14283 1.430 0.21213
## x 0.37606 0.07849 4.791 0.00492 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8307 on 5 degrees of freedom
## Multiple R-squared: 0.8211, Adjusted R-squared: 0.7854
## F-statistic: 22.96 on 1 and 5 DF, p-value: 0.004922
##
##
## $`2:-Biotin:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.06137 -0.08479 -0.10821 0.37956 -0.09211 0.27054 -0.30362
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.36851 0.26561 12.682 5.42e-05 ***
## x 0.24466 0.02466 9.921 0.000178 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.261 on 5 degrees of freedom
## Multiple R-squared: 0.9517, Adjusted R-squared: 0.942
## F-statistic: 98.42 on 1 and 5 DF, p-value: 0.0001776
##
##
## $`2:-Biotin:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.24068 -0.16024 -0.25709 -0.01833 0.16293 0.11462 -0.08257
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.10851 0.16756 24.52 2.10e-06 ***
## x 0.29515 0.01873 15.76 1.87e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1983 on 5 degrees of freedom
## Multiple R-squared: 0.9803, Adjusted R-squared: 0.9763
## F-statistic: 248.2 on 1 and 5 DF, p-value: 1.873e-05## y0 y0_lm mumax lag
## 2:-Biotin:1 98.46 5.125203 0.3760631 7.859002
## 2:-Biotin:2 96.62 29.035114 0.2446627 4.914031
## 2:-Biotin:3 107.23 60.856127 0.2951547 1.919209## 2:-Biotin:1.r2 2:-Biotin:2.r2 2:-Biotin:3.r2
## 0.8211458 0.9516543 0.9802552Specific Growth Rates Biotin trial 3 using all_easylinear:
bio3 <- fluor %>%
filter(Treatment == "-Biotin", Round %in% c(3) )%>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
bio_lin3 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, bio3, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(bio_lin3, log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(bio_lin3)## $`3:-Biotin:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.07856 -0.22592 0.15259 0.26659 0.21258 -0.23432 -0.09296
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.81040 0.35557 7.904 0.000522 ***
## x 0.23963 0.02156 11.115 0.000103 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2282 on 5 degrees of freedom
## Multiple R-squared: 0.9611, Adjusted R-squared: 0.9533
## F-statistic: 123.5 on 1 and 5 DF, p-value: 0.0001028
##
##
## $`3:-Biotin:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.2353339 -0.1112728 -0.0556536 -0.2798767 0.0004855 0.2056337 0.0053499
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.96414 0.23391 12.67 5.44e-05 ***
## x 0.26925 0.01849 14.56 2.76e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1957 on 5 degrees of freedom
## Multiple R-squared: 0.977, Adjusted R-squared: 0.9723
## F-statistic: 212 on 1 and 5 DF, p-value: 2.76e-05
##
##
## $`3:-Biotin:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.01417 -0.22670 0.78580 0.49481 -1.90075 0.22234 0.61033
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.56645 2.28724 0.248 0.814
## x 0.18591 0.09401 1.978 0.105
##
## Residual standard error: 0.9949 on 5 degrees of freedom
## Multiple R-squared: 0.4389, Adjusted R-squared: 0.3267
## F-statistic: 3.911 on 1 and 5 DF, p-value: 0.1049## y0 y0_lm mumax lag
## 3:-Biotin:1 98.82 16.616595 0.2396343 7.440080
## 3:-Biotin:2 63.40 19.378045 0.2692450 4.402396
## 3:-Biotin:3 82.43 1.761997 0.1859051 20.685290## 3:-Biotin:1.r2 3:-Biotin:2.r2 3:-Biotin:3.r2
## 0.9611016 0.9769567 0.4388915Specific Growth Rates Biotin trial 4 using all_easylinear
bio4 <- fluor %>%
filter(Treatment == "-Biotin", Round %in% c(4) )%>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
bio_lin4 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, bio4, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(bio_lin4, log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(bio_lin4)## $`4:-Biotin:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.30806 0.04318 -0.14397 -0.50617 -0.10602 0.24216 0.16276
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.26650 0.53423 2.371 0.063898 .
## x 0.23271 0.02897 8.032 0.000484 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3066 on 5 degrees of freedom
## Multiple R-squared: 0.9281, Adjusted R-squared: 0.9137
## F-statistic: 64.51 on 1 and 5 DF, p-value: 0.0004838
##
##
## $`4:-Biotin:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.19446 0.54358 -0.06320 -0.20050 -0.37665 0.05642 0.23480
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.80368 0.59222 3.046 0.02857 *
## x 0.21685 0.03212 6.752 0.00108 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3399 on 5 degrees of freedom
## Multiple R-squared: 0.9012, Adjusted R-squared: 0.8814
## F-statistic: 45.58 on 1 and 5 DF, p-value: 0.001082
##
##
## $`4:-Biotin:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.02364 0.05129 0.03152 -0.14500 -0.02402 -0.04135 0.10391
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.488502 0.120401 20.67 4.91e-06 ***
## x 0.233579 0.008269 28.25 1.04e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.08751 on 5 degrees of freedom
## Multiple R-squared: 0.9938, Adjusted R-squared: 0.9925
## F-statistic: 797.9 on 1 and 5 DF, p-value: 1.041e-06## y0 y0_lm mumax lag
## 4:-Biotin:1 57.52 3.548409 0.2327066 11.970583
## 4:-Biotin:2 67.67 6.071954 0.2168469 11.118270
## 4:-Biotin:3 97.55 12.043218 0.2335787 8.955711## 4:-Biotin:1.r2 4:-Biotin:2.r2 4:-Biotin:3.r2
## 0.9280693 0.9011555 0.9937725Specific Growth Rates Thiamine trial 1 using all_easylinear:
thia1 <- fluor %>%
filter(Treatment == "-Thiamine", Round %in% c(1) )%>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
thia_lin1 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, thia1, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(thia_lin1, log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(thia_lin1)## $`1:-Thiamine:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.30173 -0.51432 -0.04376 0.21123 0.09790 0.10677 -0.15955
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.4499 0.4668 7.391 0.000713 ***
## x 0.2417 0.0283 8.539 0.000363 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2995 on 5 degrees of freedom
## Multiple R-squared: 0.9358, Adjusted R-squared: 0.923
## F-statistic: 72.91 on 1 and 5 DF, p-value: 0.0003627
##
##
## $`1:-Thiamine:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.31890 0.23129 0.13999 -0.02326 0.25846 -0.25014 -0.03743
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.14430 0.25181 12.49 5.84e-05 ***
## x 0.26660 0.02338 11.40 9.08e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2474 on 5 degrees of freedom
## Multiple R-squared: 0.963, Adjusted R-squared: 0.9556
## F-statistic: 130 on 1 and 5 DF, p-value: 9.079e-05
##
##
## $`1:-Thiamine:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.25586 0.36063 0.14809 -0.29385 0.02147 -0.02179 0.04130
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.33255 0.25195 13.23 4.41e-05 ***
## x 0.26006 0.02339 11.12 0.000103 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2476 on 5 degrees of freedom
## Multiple R-squared: 0.9611, Adjusted R-squared: 0.9533
## F-statistic: 123.6 on 1 and 5 DF, p-value: 0.0001027## y0 y0_lm mumax lag
## 1:-Thiamine:1 173.20 31.49697 0.2416543 7.0536952
## 1:-Thiamine:2 44.15 23.20350 0.2666038 2.4129054
## 1:-Thiamine:3 33.19 28.00966 0.2600576 0.6525443## 1:-Thiamine:1.r2 1:-Thiamine:2.r2 1:-Thiamine:3.r2
## 0.9358222 0.9629722 0.9611167Specific Growth Rates Thiamine trial 2 using all_easylinear
thia2 <- fluor %>%
filter(Treatment == "-Thiamine", Round %in% c(2) )%>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
thia_lin2 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, thia2, h = 9, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(thia_lin2, log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(thia_lin2)## $`2:-Thiamine:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.4806 -0.2537 -0.1203 0.2472 0.5625
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.63297 0.26430 17.53 4.84e-07 ***
## x 0.24127 0.02348 10.27 1.79e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3638 on 7 degrees of freedom
## Multiple R-squared: 0.9378, Adjusted R-squared: 0.9289
## F-statistic: 105.5 on 1 and 7 DF, p-value: 1.79e-05
##
##
## $`2:-Thiamine:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.35944 -0.06839 0.00212 0.08203 0.24602
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.80494 0.11101 43.28 9.17e-10 ***
## x 0.22326 0.01166 19.15 2.64e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1806 on 7 degrees of freedom
## Multiple R-squared: 0.9813, Adjusted R-squared: 0.9786
## F-statistic: 366.7 on 1 and 7 DF, p-value: 2.636e-07
##
##
## $`2:-Thiamine:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.42503 -0.12665 -0.00432 0.16254 0.40225
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.54004 0.16251 27.94 1.93e-08 ***
## x 0.25614 0.01707 15.01 1.40e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2644 on 7 degrees of freedom
## Multiple R-squared: 0.9699, Adjusted R-squared: 0.9656
## F-statistic: 225.2 on 1 and 7 DF, p-value: 1.4e-06## y0 y0_lm mumax lag
## 2:-Thiamine:1 156.08 102.81917 0.2412673 1.7300188
## 2:-Thiamine:2 131.83 122.11205 0.2232630 0.3429772
## 2:-Thiamine:3 140.09 93.69448 0.2561405 1.5704110## 2:-Thiamine:1.r2 2:-Thiamine:2.r2 2:-Thiamine:3.r2
## 0.9378050 0.9812701 0.9698585Specific Growth Rates Thiamine trial 3 using all_easylinear:
thia3 <- fluor %>%
filter(Treatment == "-Thiamine", Round %in% c(3) )%>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
thia_lin3 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, thia3, h = 9, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(thia_lin3, log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(thia_lin3)## $`3:-Thiamine:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.35661 -0.21370 -0.07224 0.18665 0.48952
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.79058 0.36271 4.937 0.00168 **
## x 0.25375 0.01937 13.101 3.52e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3001 on 7 degrees of freedom
## Multiple R-squared: 0.9608, Adjusted R-squared: 0.9552
## F-statistic: 171.6 on 1 and 7 DF, p-value: 3.521e-06
##
##
## $`3:-Thiamine:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.49865 0.01803 0.07384 0.10434 0.40557
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.81256 0.35153 5.156 0.00132 **
## x 0.31014 0.02091 14.833 1.52e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3239 on 7 degrees of freedom
## Multiple R-squared: 0.9692, Adjusted R-squared: 0.9648
## F-statistic: 220 on 1 and 7 DF, p-value: 1.517e-06
##
##
## $`3:-Thiamine:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.4806 -0.1946 -0.0558 0.3160 0.5117
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.99093 0.40297 4.941 0.00167 **
## x 0.25389 0.02397 10.593 1.46e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3713 on 7 degrees of freedom
## Multiple R-squared: 0.9413, Adjusted R-squared: 0.9329
## F-statistic: 112.2 on 1 and 7 DF, p-value: 1.462e-05## y0 y0_lm mumax lag
## 3:-Thiamine:1 49.92 5.992945 0.2537505 8.354027
## 3:-Thiamine:2 72.28 6.126119 0.3101377 7.957711
## 3:-Thiamine:3 57.00 7.322315 0.2538908 8.082708## 3:-Thiamine:1.r2 3:-Thiamine:2.r2 3:-Thiamine:3.r2
## 0.9608132 0.9691648 0.9412777Specific Growth Rates Thiamine trial 4 using all_easylinear
thia4 <- fluor %>%
filter(Treatment == "-Thiamine", Round %in% c(4) )%>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
thia_lin4 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, thia4, h = 9, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(thia_lin4, log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(thia_lin4)## $`4:-Thiamine:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.34954 -0.14112 0.08196 0.17775 0.22950
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.4689 0.1816 19.11 2.68e-07 ***
## x 0.2533 0.0139 18.22 3.71e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2153 on 7 degrees of freedom
## Multiple R-squared: 0.9794, Adjusted R-squared: 0.9764
## F-statistic: 332 on 1 and 7 DF, p-value: 3.71e-07
##
##
## $`4:-Thiamine:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.269152 -0.100510 0.007589 0.106550 0.281775
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.87449 0.21124 13.61 2.72e-06 ***
## x 0.24309 0.01256 19.35 2.46e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1946 on 7 degrees of freedom
## Multiple R-squared: 0.9816, Adjusted R-squared: 0.979
## F-statistic: 374.3 on 1 and 7 DF, p-value: 2.457e-07
##
##
## $`4:-Thiamine:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.7143 0.0431 0.1993 0.3349 0.4359
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.65854 0.76560 3.472 0.0104 *
## x 0.23859 0.04554 5.239 0.0012 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7055 on 7 degrees of freedom
## Multiple R-squared: 0.7968, Adjusted R-squared: 0.7678
## F-statistic: 27.45 on 1 and 7 DF, p-value: 0.0012## y0 y0_lm mumax lag
## 4:-Thiamine:1 161.92 32.10143 0.2532588 6.389518
## 4:-Thiamine:2 76.34 17.71644 0.2430897 6.008910
## 4:-Thiamine:3 109.65 14.27538 0.2385910 8.544988## 4:-Thiamine:1.r2 4:-Thiamine:2.r2 4:-Thiamine:3.r2
## 0.9793538 0.9816432 0.7968191Specific Growth Rates Replete trial 1 using all_easylinear:
rep1 <- fluor %>%
filter(Treatment == "Replete", Round %in% c(1)) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
rep_lin1 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, rep1, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(rep_lin1, log = "y")
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(rep_lin1)## $`1:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.23658 0.13397 0.19451 -0.01732 -0.04291 0.10941 -0.14107
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.29023 0.23491 14.01 3.34e-05 ***
## x 0.21573 0.01613 13.37 4.19e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1707 on 5 degrees of freedom
## Multiple R-squared: 0.9728, Adjusted R-squared: 0.9674
## F-statistic: 178.8 on 1 and 5 DF, p-value: 4.186e-05
##
##
## $`1:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.01471 0.09138 -0.04521 -0.16653 0.07629 0.15920 -0.10042
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.03875 0.17358 17.51 1.12e-05 ***
## x 0.22585 0.01192 18.95 7.55e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1262 on 5 degrees of freedom
## Multiple R-squared: 0.9863, Adjusted R-squared: 0.9835
## F-statistic: 358.9 on 1 and 5 DF, p-value: 7.551e-06
##
##
## $`1:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.025927 0.009719 0.219121 -0.438641 0.040203 0.154873 -0.011203
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.29711 0.31788 10.37 0.000143 ***
## x 0.22019 0.02183 10.09 0.000164 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.231 on 5 degrees of freedom
## Multiple R-squared: 0.9531, Adjusted R-squared: 0.9438
## F-statistic: 101.7 on 1 and 5 DF, p-value: 0.0001641## y0 y0_lm mumax lag
## 1:Replete:1 66.01 26.84907 0.2157266 4.169977
## 1:Replete:2 87.82 20.87905 0.2258538 6.360502
## 1:Replete:3 94.03 27.03446 0.2201909 5.661004## 1:Replete:1.r2 1:Replete:2.r2 1:Replete:3.r2
## 0.9727947 0.9862601 0.9531494Specific Growth Rates Replete trial 2 using all_easylinear:
rep2 <- fluor %>%
filter(Treatment == "Replete", Round %in% c(2)) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
rep_lin2 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, rep2, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(rep_lin2, log = "y")
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(rep_lin2)## $`2:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.18529 -0.03780 -0.16478 -0.18114 0.17936 -0.07894 0.09801
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.37352 0.11553 37.86 2.42e-07 ***
## x 0.18726 0.01602 11.69 8.05e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1695 on 5 degrees of freedom
## Multiple R-squared: 0.9647, Adjusted R-squared: 0.9576
## F-statistic: 136.6 on 1 and 5 DF, p-value: 8.053e-05
##
##
## $`2:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.04068 0.56895 -0.48634 -0.31279 -0.05303 0.38914 -0.06524
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.36733 0.27564 15.844 1.82e-05 ***
## x 0.20689 0.03822 5.412 0.00291 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4045 on 5 degrees of freedom
## Multiple R-squared: 0.8542, Adjusted R-squared: 0.825
## F-statistic: 29.29 on 1 and 5 DF, p-value: 0.002913
##
##
## $`2:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.17429 0.11485 0.12945 0.04594 -0.14406 0.10395 -0.07585
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.38836 0.22011 15.39 2.1e-05 ***
## x 0.24033 0.01335 18.01 9.7e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1412 on 5 degrees of freedom
## Multiple R-squared: 0.9848, Adjusted R-squared: 0.9818
## F-statistic: 324.3 on 1 and 5 DF, p-value: 9.701e-06## y0 y0_lm mumax lag
## 2:Replete:1 95.47 79.32272 0.1872621 0.9894552
## 2:Replete:2 75.69 78.83269 0.2068881 -0.1966365
## 2:Replete:3 51.77 29.61731 0.2403281 2.3237062## 2:Replete:1.r2 2:Replete:2.r2 2:Replete:3.r2
## 0.9646965 0.8542034 0.9848147Specific Growth Rates Replete trial 3 using all_easylinear
rep3 <- fluor %>%
filter(Treatment == "Replete", Round %in% c(3)) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
rep_lin3 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, rep3, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(rep_lin3, log = "y")
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(rep_lin3)## $`3:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.056275 0.286364 -0.388386 0.001474 -0.003777 -0.212798 0.260848
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.26798 0.31594 7.179 0.000816 ***
## x 0.26516 0.02498 10.616 0.000128 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2643 on 5 degrees of freedom
## Multiple R-squared: 0.9575, Adjusted R-squared: 0.949
## F-statistic: 112.7 on 1 and 5 DF, p-value: 0.0001282
##
##
## $`3:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.01109 -0.07478 -0.04567 0.22757 -0.09197 -0.00871 -0.01753
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.69027 0.20383 8.292 0.000416 ***
## x 0.25682 0.01105 23.232 2.75e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.117 on 5 degrees of freedom
## Multiple R-squared: 0.9908, Adjusted R-squared: 0.989
## F-statistic: 539.7 on 1 and 5 DF, p-value: 2.75e-06
##
##
## $`3:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.37438 -0.38671 -0.12656 -0.27670 0.47488 0.07385 -0.13314
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.09308 0.42459 4.93 0.004361 **
## x 0.30144 0.03357 8.98 0.000286 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3552 on 5 degrees of freedom
## Multiple R-squared: 0.9416, Adjusted R-squared: 0.9299
## F-statistic: 80.65 on 1 and 5 DF, p-value: 0.0002856## y0 y0_lm mumax lag
## 3:Replete:1 46.52 9.659839 0.2651609 5.928118
## 3:Replete:2 77.27 5.420959 0.2568156 10.346075
## 3:Replete:3 61.72 8.109824 0.3014418 6.732749## 3:Replete:1.r2 3:Replete:2.r2 3:Replete:3.r2
## 0.9575206 0.9908211 0.9416198Specific Growth Rates Replete trial 4 using all_easylinear:
rep4 <- fluor %>%
filter(Treatment == "Replete", Round %in% c(4)) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
rep_lin4 <- all_easylinear(Chl ~ Time|Round+Treatment+Rep, rep4, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(rep_lin4, log = "y")
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(rep_lin4)## $`4:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.09195 0.19036 0.07953 -0.20968 -0.15320 0.21726 -0.03232
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.00947 0.25121 11.98 7.15e-05 ***
## x 0.21058 0.01725 12.21 6.53e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1826 on 5 degrees of freedom
## Multiple R-squared: 0.9675, Adjusted R-squared: 0.961
## F-statistic: 149 on 1 and 5 DF, p-value: 6.529e-05
##
##
## $`4:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.11124 0.04917 -0.34079 0.14249 -0.10230 0.22701 -0.08682
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.08172 0.28910 10.66 0.000126 ***
## x 0.20366 0.01986 10.26 0.000151 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2101 on 5 degrees of freedom
## Multiple R-squared: 0.9546, Adjusted R-squared: 0.9456
## F-statistic: 105.2 on 1 and 5 DF, p-value: 0.0001513
##
##
## $`4:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.14336 0.08611 -0.35495 0.21552 -0.30373 0.09004 0.12366
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.07289 0.43805 4.732 0.005187 **
## x 0.17495 0.02376 7.364 0.000725 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2514 on 5 degrees of freedom
## Multiple R-squared: 0.9156, Adjusted R-squared: 0.8987
## F-statistic: 54.23 on 1 and 5 DF, p-value: 0.0007253## y0 y0_lm mumax lag
## 4:Replete:1 78.16 20.276716 0.2105809 6.407442
## 4:Replete:2 157.37 21.795906 0.2036571 9.706892
## 4:Replete:3 87.34 7.947773 0.1749477 13.700762## 4:Replete:1.r2 4:Replete:2.r2 4:Replete:3.r2
## 0.9675256 0.9546300 0.91558372.4.1 Combine all specific growth rate values from each treatment:
extract_results <- function(model, treatment_name) {
coefs <- coef(model)
r2vals <- rsquared(model)
df <- as.data.frame(coefs) %>%
tibble::rownames_to_column("Group") %>%
mutate(
Treatment = treatment_name,
r2 = r2vals
) %>%
# Split "Group" into Round and Rep (Group looks like "1:-Thiamine:2")
tidyr::separate(Group, into = c("Round", "Treatment_label", "Rep"), sep = ":", remove = FALSE) %>%
mutate(Round = as.integer(Round))
return(df)
}
# --- Extract Biotin rounds ---
df_bio1 <- extract_results(bio_lin1, "-Biotin")
df_bio2 <- extract_results(bio_lin2, "-Biotin")
df_bio3 <- extract_results(bio_lin3, "-Biotin")
df_bio4 <- extract_results(bio_lin4, "-Biotin")
df_bio <- bind_rows(df_bio1, df_bio2, df_bio3, df_bio4)
# --- Extract Thiamine rounds ---
df_thia1 <- extract_results(thia_lin1, "-Thiamine")
df_thia2 <- extract_results(thia_lin2, "-Thiamine")
df_thia3 <- extract_results(thia_lin3, "-Thiamine")
df_thia4 <- extract_results(thia_lin4, "-Thiamine")
df_thia <- bind_rows(df_thia1, df_thia2, df_thia3, df_thia4)
# --- Extract Replete rounds ---
df_rep1 <- extract_results(rep_lin1, "Replete")
df_rep2 <- extract_results(rep_lin2, "Replete")
df_rep3 <- extract_results(rep_lin3, "Replete")
df_rep4 <- extract_results(rep_lin4, "Replete")
df_rep <- bind_rows(df_rep1, df_rep2, df_rep3, df_rep4)
# --- Combine all treatments ---
all_lin_results <- bind_rows(df_bio, df_thia, df_rep)
# Optional: ensure Round is treated as a factor
all_lin_results$Round <- factor(all_lin_results$Round)2.4.2 Run an ANOVA on specific growth rate values and Tukey HSD post-hoc test for treatment:
all_lin_results$Round <- factor(all_lin_results$Round)
linear <- aov(mumax ~ Treatment , data = all_lin_results)
summary(linear)## Df Sum Sq Mean Sq F value Pr(>F)
## Treatment 2 0.00517 0.002586 1.861 0.172
## Residuals 33 0.04587 0.001390## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = mumax ~ Treatment, data = all_lin_results)
##
## $Treatment
## diff lwr upr p adj
## -Thiamine--Biotin 0.005533672 -0.03181359 0.042880937 0.9298833
## Replete--Biotin -0.022203946 -0.05955121 0.015143319 0.3234309
## Replete--Thiamine -0.027737618 -0.06508488 0.009609647 0.17802432.4.3 Plot specific growth rates for Pyro grown in -Thiamine, -Biotin, and Replete:
all_lin_results$Treatment <- factor(
all_lin_results$Treatment,
levels = c("-Cobalamin", "-Thiamine", "-Biotin","Replete")
)
mu <- ggplot(all_lin_results, aes(x = Treatment, y = mumax, color = Treatment)) +
# Raw points
geom_point(size = 3, shape = 16, position = position_jitter(width = 0.1)) +
# Mean ± SE
stat_summary(fun = mean, geom = "point", shape = 1, size = 5, stroke = 1.2) +
stat_summary(fun.data = mean_se, geom = "errorbar", width = 0.2) +
scale_color_manual(values = c(
"-Cobalamin" = "#102E50",
"Replete" = "#F5C45E",
"-Thiamine" = "deeppink4",
"-Biotin" = "#667028"
)) +
scale_x_discrete(labels = c(
"-Cobalamin" = expression("\u2212B"[12]),
"Replete" = "Replete",
"-Thiamine" = expression("\u2212Thiamine"),
"-Biotin" = expression("\u2212Biotin")
)) +
theme_minimal() +
theme(
legend.position = "none",
panel.grid = element_blank(),
panel.border = element_rect(fill = NA, color = "black"),
axis.title = element_text( size = 20),
axis.text = element_text( size = 20, color = "black"),
axis.text.x = element_text(angle = 45, hjust = 1),
strip.text = element_text( size = 20, color = "black"),
legend.title = element_blank(),
legend.text = element_blank(),
legend.key = element_blank()
) +
labs(
x = "Media Treatment",
y = expression(paste("(", mu, ") RFU" * "\u00B7" * "day"^-1, " "))
) + guides(color = "none", fill = "none", shape = "none")
mu
2.5 Figure 3
bo <-
LD_OTB_all_growth /
(max_bio_LD +
mu ) +
plot_layout(guides = "collect") &
theme(
legend.position = "bottom",
panel.spacing = unit(.1, "lines"),
plot.margin = margin(15, 15, 15, 15),
axis.title.y = element_text(margin = margin(r = 12)),
axis.title.x = element_text(margin = margin(t = 8)),
legend.margin = margin(t = 4, b = 4),
legend.box.margin = margin(5, 5, 5, 5),
axis.text.x = element_text(size = 16),
)
bo
Fig. 3 Growth responses of P. bahamense when grown in replete L1/4 media and in media missing vitamin B12, thiamine, or biotin. A In vivo Chlorophyll a fluorescence (Chl-a) was used to track P. bahamense growth across four experimental rounds. The replicates growing in replete media and in media missing thiamine or biotin were serially transferred between each round. Since replicates in media without B12 failed to grow, the replete replicates were used to restart the cultures in media missing B12 in each round. B Maximum Chl-a values in each replicate by media formulation. All data are plotted with color representing the media treatment. Means are plotted as black open circles, and error bars represent one standard deviation of the mean. ANOVA results indicate media had a significant effect on maximum Chl-a (p = 0.0000, F = 45.35). Post-hoc multiple comparisons using the Tukey HSD test indicate that the maximum biomasses achieved in replicates grown in media missing B12 were significantly lower than in replicates grown in replete media (p-adj = 0.0001), media missing thiamine (p-adj = 0.0001), and media missing biotin (p-adj = 0.0001). Replicates grown in media missing thiamine reached significantly higher Chl-a values than replicates grown in replete media (p-adj = 0.0001) or in media without biotin (p-adj = 0.001). The maximum Chl-a values for replicates grown in media missing biotin were not significantly different from replicates in replete media (p-adj = 0.369). Results from comparisons to replete are shown on the plot with the following significance codes: p-adj > 0.05; ‘n.s.’ and p-adj < 0.001; ’*’. C** Specific growth rate (µ) of each replicate in –thiamine, –biotin, and replete L1/4 media. Growth rates could not be calculated for replicates in L1/4 –B12 due to low and variable Chl-a values. Means are plotted as black open circles and error bars represent one standard deviation of the mean. Excluding the –B12 treatment, media treatment did not significantly affect specific growth rate (one-way ANOVA; p = 0.172, F = 1.861).
3 Evaluating the B12 requirements of P. bahamense
Chlorophyll-a fluorescence (RFU) data from P. bahamense cultures were used to visualize growth patterns, quantify maximum biomass, and estimate specific growth rates under media conditions deficient in media lacking one targeted B vitamin. Specific growth rates were computed using the “alleasy_linear” function of the “growthrates” package in R Studio (Hall et al., 2014; RStudio Team, 2020).
3.1 Import and Parse the Dataset:
b12 <-as.data.frame(read_csv("b12_levels.csv"))
# Remove periods and convert to numeric
b12$Time <- as.numeric(sub("T", "", b12$Time))
# Convert Date column to Date type
b12$Date <- mdy(b12$Date)
b12$Treatment <- unlist(b12$Treatment)
#covert timepoints to hours
b12$Time <- b12$Time * 48 # hours
b12$Time <- b12$Time / 24
# put treatment level in desired order
b12 <- b12 %>%
mutate(Treatment = factor(Treatment, levels = c(
"369pM", # Place the levels in desired order
"36pM",
"3.69pM",
"0.369pM",
"0.0369pM"
)))3.2 Prepare data for plotting:
3.3 Visulaize growth
growth <- ggplot(b12, aes(x = Time, y = `Chlorophyll-A (RFU)`, color = Treatment)) +
geom_point() +
geom_smooth(aes(color = Treatment), se = FALSE) +
facet_grid(Round ~ Treatment) +
labs(
x = "Time (days)",
y = expression(paste("Chlorophyll-", italic("a"), " fluorescence (RFU)"))
) +
scale_y_continuous(labels = scales::comma) +
scale_color_manual(values = b12_palette) +
theme_minimal() +
theme(
panel.border = element_rect(fill = NA, color = "black"),
panel.grid = element_blank(),
axis.text.y = element_text(size = 20, color = "black"),
axis.text.x = element_text(size = 12, color = "black"),
axis.title.y = element_text(size = 20, margin = margin(t = 10)),
axis.title.x = element_text(size = 20, margin = margin(t = 10)),
strip.text.x = element_text(size = 16),
strip.text = element_text( size = 16, color = "black"),
legend.title = element_text( size = 16, color = "black"),
legend.text = element_text( size = 16, color = "black"),
legend.position = "bottom",
axis.ticks = element_line(color = "black", linewidth = 0.5),
axis.ticks.length = unit(4, "pt"),
panel.spacing = unit(.15, "cm")
)
growth
3.4 Calculate and visualize maximum RFU values for each round and medium treatment:
3.4.1 Calculate maximum RFU:
max_bio <- b12 %>%
filter(Treatment %in% c(Treatment, levels = c("369pM", "36pM", "3.69pM", "0.369pM", "0.0369pM"))) %>%
group_by(Treatment, Bio_Rep) %>%
summarise(Max_Chl = max(`Chlorophyll-A (RFU)`, na.rm = TRUE)) %>%
ungroup()
print(max_bio)## # A tibble: 15 × 3
## Treatment Bio_Rep Max_Chl
## <fct> <dbl> <dbl>
## 1 0.0369pM 1 842.
## 2 0.0369pM 2 541.
## 3 0.0369pM 3 928.
## 4 0.369pM 1 1968.
## 5 0.369pM 2 1897.
## 6 0.369pM 3 1866.
## 7 3.69pM 1 1923.
## 8 3.69pM 2 3929.
## 9 3.69pM 3 2146.
## 10 36pM 1 5580.
## 11 36pM 2 6041.
## 12 36pM 3 4801.
## 13 369pM 1 5287.
## 14 369pM 2 4424.
## 15 369pM 3 5539.3.4.2 Run an ANOVA followed by a Tukey HSD test for the treatment factor:
## Df Sum Sq Mean Sq F value Pr(>F)
## Treatment 4 49705730 12426432 31.27 1.25e-05 ***
## Residuals 10 3973244 397324
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Max_Chl ~ Treatment, data = max_bio)
##
## $Treatment
## diff lwr upr p adj
## 0.369pM-0.0369pM 1139.8033 -554.0107 2833.617 0.2494204
## 3.69pM-0.0369pM 1895.5967 201.7826 3589.411 0.0272120
## 36pM-0.0369pM 4703.7467 3009.9326 6397.561 0.0000277
## 369pM-0.0369pM 4312.9433 2619.1293 6006.757 0.0000597
## 3.69pM-0.369pM 755.7933 -938.0207 2449.607 0.6023509
## 36pM-0.369pM 3563.9433 1870.1293 5257.757 0.0003042
## 369pM-0.369pM 3173.1400 1479.3259 4866.954 0.0007806
## 36pM-3.69pM 2808.1500 1114.3359 4501.964 0.0020036
## 369pM-3.69pM 2417.3467 723.5326 4111.161 0.0058828
## 369pM-36pM -390.8033 -2084.6174 1303.011 0.93665083.4.3 Plot maximum biomass:
# Define factor levels (low to high B12)
levels_order <- c("0.0369pM", "0.369pM", "3.69pM", "36pM", "369pM")
max_bio$Treatment <- factor(trimws(max_bio$Treatment), levels = levels_order)
# Calculate summary statistics
summary_stats <- max_bio %>%
group_by(Treatment) %>%
summarise(
mean_chl = mean(Max_Chl, na.rm = TRUE),
sd_chl = sd(Max_Chl, na.rm = TRUE),
.groups = "drop"
)
summary_stats$Treatment <- factor(summary_stats$Treatment, levels = levels_order)
# Define color palette
b12_palette <- c(
"0.0369pM" = "lightblue",
"0.369pM" = "#99c2e6",
"3.69pM" = "#66a3d2",
"36pM" = "#3380bf",
"369pM" = "#004080"
)
max_bio_plot <- ggplot(summary_stats, aes(x = Treatment)) +
# Replicate points
geom_point(
data = max_bio,
aes(y = Max_Chl, color = Treatment),
size = 3,
alpha = 1,
position = position_jitter(width = 0.1)
) +
stat_summary(
data = max_bio,
aes(y = Max_Chl, color = Treatment),
fun = mean,
geom = "point",
shape = 1,
size = 5,
stroke = 1.2
) +
stat_summary(
data = max_bio,
aes(y = Max_Chl, color = Treatment),
fun.data = mean_sdl,
fun.args = list(mult = 1),
geom = "errorbar",
width = 0.2
)+
# Keep your palette but remove the scale (so no legend)
scale_color_manual(values = b12_palette, guide = "none") +
scale_y_continuous(labels = scales::comma) +
coord_cartesian(ylim = c(0, 10000)) +
ylab(expression(paste("Maximum RFU"))) +
xlab(expression(B[12]~Concentration~"(pM)")) +
theme_minimal() +
theme(
panel.border = element_rect(fill = NA, color = "black"),
panel.grid = element_blank(),
axis.text.y = element_text(size = 20, color = "black"),
axis.text.x = element_text(size = 20, color = "black", angle = 45, hjust = 1),
axis.title.y = element_text(size = 20),
axis.title.x = element_text(size = 20),
legend.position = "none",
axis.ticks = element_line(color = "black", linewidth = 0.5),
axis.ticks.length = unit(4, "pt")
)
max_bio_plot
3.5 Specific growth rates using all_easylinear:
3.5.1 Remove outliers:
# ️⃣ Compute T0 baseline per replicate, treatment, and round
baseline <- aggregate(
`Chlorophyll-A (RFU)` ~ Bio_Rep + Treatment + Round,
data = subset(b12, Time == 0),
FUN = mean,
na.rm = TRUE
)
names(baseline)[names(baseline) == "Chlorophyll-A (RFU)"] <- "Chl_T0"
# Merge baseline back into the full dataset
b12_merged <- merge(b12, baseline, by = c("Bio_Rep", "Treatment", "Round"), all.x = TRUE)
# Keep all T0 points, and remove any post-T0 points where replicate < its T0
b12_clean <- subset(
b12_merged,
Time == 0 | `Chlorophyll-A (RFU)` >= Chl_T0
)Specific growth rate with all easy linear for 0.0369pM:
pM_1 <- b12_clean %>%
filter(
Treatment %in% c("0.0369pM"),
Round %in% c(1)
) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
mu1 <- all_easylinear(Chl ~ Time | Round + Treatment + Bio_Rep, pM_1, h =4, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(mu1, log = "y")
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(mu1)## $`1:0.0369pM:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## 0.21354 -0.37851 0.11642 0.04856
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.72824 0.52579 8.993 0.0121 *
## x 0.19590 0.07155 2.738 0.1115
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.32 on 2 degrees of freedom
## Multiple R-squared: 0.7894, Adjusted R-squared: 0.6841
## F-statistic: 7.496 on 1 and 2 DF, p-value: 0.1115
##
##
## $`1:0.0369pM:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## -0.10998 0.29391 -0.05595 -0.12799
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.39799 0.22595 23.890 0.00175 **
## x 0.07310 0.02382 3.069 0.09177 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2429 on 2 degrees of freedom
## Multiple R-squared: 0.8249, Adjusted R-squared: 0.7373
## F-statistic: 9.421 on 1 and 2 DF, p-value: 0.09177
##
##
## $`1:0.0369pM:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## 0.29094 -0.55228 0.23175 0.02959
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.5299 0.5772 7.848 0.0159 *
## x 0.2125 0.1054 2.017 0.1813
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4713 on 2 degrees of freedom
## Multiple R-squared: 0.6703, Adjusted R-squared: 0.5055
## F-statistic: 4.067 on 1 and 2 DF, p-value: 0.1813## y0 y0_lm mumax lag
## 1:0.0369pM:1 145.02 113.09593 0.19590188 1.2691828
## 1:0.0369pM:2 197.95 220.96199 0.07310236 -1.5044141
## 1:0.0369pM:3 97.44 92.75275 0.21252290 0.2319725## 1:0.0369pM:1.r2 1:0.0369pM:2.r2 1:0.0369pM:3.r2
## 0.7893916 0.8248814 0.6703408Specific growth rate with all easy linear for 0.369pM:
pM_2 <- b12_clean %>%
filter(Treatment %in% c("0.369pM"),
Round %in% c(1),
Time >= 0) %>% # <-- exclude T0
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
mu2 <-all_easylinear(Chl ~ Time|Treatment+Bio_Rep, pM_2, h = 4, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(mu2, log = "y") # line only
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(mu2)## $`0.369pM:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## -0.03871 0.14944 -0.14406 0.03332
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.72896 0.22995 20.566 0.00236 **
## x 0.23253 0.02555 9.101 0.01186 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1512 on 2 degrees of freedom
## Multiple R-squared: 0.9764, Adjusted R-squared: 0.9646
## F-statistic: 82.83 on 1 and 2 DF, p-value: 0.01186
##
##
## $`0.369pM:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## -0.03225 -0.15070 0.39814 -0.21519
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.12741 0.70102 7.314 0.0182 *
## x 0.20224 0.07559 2.675 0.1159
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3381 on 2 degrees of freedom
## Multiple R-squared: 0.7816, Adjusted R-squared: 0.6724
## F-statistic: 7.157 on 1 and 2 DF, p-value: 0.1159
##
##
## $`0.369pM:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## -0.26163 0.40004 -0.01518 -0.12322
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.21953 0.72414 5.827 0.0282 *
## x 0.28628 0.07809 3.666 0.0670 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3492 on 2 degrees of freedom
## Multiple R-squared: 0.8705, Adjusted R-squared: 0.8057
## F-statistic: 13.44 on 1 and 2 DF, p-value: 0.06701## y0 y0_lm mumax lag
## 0.369pM:1 248.73 113.17754 0.2325283 3.386298
## 0.369pM:2 375.27 168.57963 0.2022351 3.956966
## 0.369pM:3 182.06 68.00143 0.2862760 3.440063## 0.369pM:1.r2 0.369pM:2.r2 0.369pM:3.r2
## 0.9764234 0.7815961 0.8704739Specific growth rate with all easy linear for 3.69pM:
pM_3 <- b12_clean %>%
filter(Treatment %in% c("3.69pM"),
Round %in% c(1),
Time >= 0) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
mu3 <-all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, pM_3, h =5, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(mu3 , log = "y")
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(mu3)## $`1:3.69pM:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.03126 -0.21999 0.32699 0.13101 -0.20676
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.60153 0.36017 12.776 0.00103 **
## x 0.24850 0.04245 5.854 0.00994 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2685 on 3 degrees of freedom
## Multiple R-squared: 0.9195, Adjusted R-squared: 0.8927
## F-statistic: 34.28 on 1 and 3 DF, p-value: 0.009935
##
##
## $`1:3.69pM:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.100883 -0.081917 -0.005946 -0.145891 0.132871
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.40262 0.10570 51.11 1.65e-05 ***
## x 0.21586 0.02158 10.01 0.00213 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1365 on 3 degrees of freedom
## Multiple R-squared: 0.9709, Adjusted R-squared: 0.9612
## F-statistic: 100.1 on 1 and 3 DF, p-value: 0.002126
##
##
## $`1:3.69pM:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.076428 0.082299 0.033975 -0.009133 -0.030712
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.8058 0.1154 41.65 3.05e-05 ***
## x 0.2069 0.0111 18.63 0.000337 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.07023 on 3 degrees of freedom
## Multiple R-squared: 0.9914, Adjusted R-squared: 0.9886
## F-statistic: 347.1 on 1 and 3 DF, p-value: 0.0003375## y0 y0_lm mumax lag
## 1:3.69pM:1 142.16 99.63649 0.2485048 1.4302532
## 1:3.69pM:2 245.55 221.98664 0.2158566 0.4673628
## 1:3.69pM:3 271.24 122.21605 0.2068770 3.8535635## 1:3.69pM:1.r2 1:3.69pM:2.r2 1:3.69pM:3.r2
## 0.9195173 0.9708995 0.9914315Specific growth rate with all easy linear for 36pM:
pM_4 <- b12_clean %>%
filter(Treatment %in% c("36pM"),
Round %in% c(1),
Time >= 0) %>% # <-- exclude T0
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
mu4 <-all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, pM_4, h = 5, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(mu4 , log = "y")
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(mu4)## $`1:36pM:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.02088 -0.04810 0.12932 -0.19786 0.09576
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.69402 0.24700 19.00 0.000318 ***
## x 0.23918 0.02377 10.06 0.002090 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1503 on 3 degrees of freedom
## Multiple R-squared: 0.9712, Adjusted R-squared: 0.9616
## F-statistic: 101.3 on 1 and 3 DF, p-value: 0.00209
##
##
## $`1:36pM:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.5827 0.4830 0.2809 0.1390 -0.3201
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.92981 0.81583 6.043 0.00909 **
## x 0.18334 0.06122 2.995 0.05791 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5078 on 3 degrees of freedom
## Multiple R-squared: 0.7494, Adjusted R-squared: 0.6658
## F-statistic: 8.969 on 1 and 3 DF, p-value: 0.05791
##
##
## $`1:36pM:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.05819 0.08475 -0.32792 0.16882 0.01616
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.22326 0.36356 11.617 0.00137 **
## x 0.27122 0.03498 7.753 0.00446 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2213 on 3 degrees of freedom
## Multiple R-squared: 0.9525, Adjusted R-squared: 0.9366
## F-statistic: 60.11 on 1 and 3 DF, p-value: 0.004463## y0 y0_lm mumax lag
## 1:36pM:1 330.46 109.29127 0.2391784 4.626125
## 1:36pM:2 205.58 138.35322 0.1833369 2.160096
## 1:36pM:3 104.26 68.25567 0.2712184 1.561941## 1:36pM:1.r2 1:36pM:2.r2 1:36pM:3.r2
## 0.9712271 0.7493526 0.9524610Specific growth rate with all easy linear for 369pM:
pM_5 <- b12_clean %>%
filter(Treatment %in% c("369pM"),
Round %in% c(1),
Time >= 0) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
mu5 <-all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, pM_5, h = 5, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(mu5 , log = "y")
# Shows the full regression output: coefficients, SEs, t-tests, p-values, and model fit statistics
summary(mu5)## $`1:369pM:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.08633 0.17016 -0.04861 0.01842 -0.05363
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.74188 0.15680 30.24 7.94e-05 ***
## x 0.25385 0.01538 16.51 0.000484 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1183 on 3 degrees of freedom
## Multiple R-squared: 0.9891, Adjusted R-squared: 0.9855
## F-statistic: 272.6 on 1 and 3 DF, p-value: 0.0004836
##
##
## $`1:369pM:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.221300 -0.358207 0.001927 -0.035733 0.170713
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.6394 0.3488 13.303 0.000918 ***
## x 0.2400 0.0342 7.017 0.005946 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2631 on 3 degrees of freedom
## Multiple R-squared: 0.9426, Adjusted R-squared: 0.9234
## F-statistic: 49.23 on 1 and 3 DF, p-value: 0.005946
##
##
## $`1:369pM:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.26764 0.51741 -0.14472 -0.19223 0.08718
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.27559 0.49285 8.675 0.00322 **
## x 0.25738 0.05808 4.431 0.02135 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3673 on 3 degrees of freedom
## Multiple R-squared: 0.8675, Adjusted R-squared: 0.8233
## F-statistic: 19.64 on 1 and 3 DF, p-value: 0.02135## y0 y0_lm mumax lag
## 1:369pM:1 286.00 114.64947 0.2538508 3.600983
## 1:369pM:2 321.18 103.48174 0.2399582 4.720016
## 1:369pM:3 106.96 71.92225 0.2573820 1.541947## 1:369pM:1.r2 1:369pM:2.r2 1:369pM:3.r2
## 0.9891142 0.9425652 0.86746863.5.2 Make dataframe with fits:
make_df <- function(model, round_num, experiment_name) {
coefs <- coef(model)
if (is.matrix(coefs)) {
df <- as.data.frame(coefs) %>%
tibble::rownames_to_column("Group") %>%
mutate(
Round = round_num,
Experiment = experiment_name,
r2 = rsquared(model)
)
} else {
df <- data.frame(
Group = names(coefs),
Estimate = as.numeric(coefs),
Round = round_num,
Experiment = experiment_name,
r2 = rep(rsquared(model), length(coefs))
)
}
return(df)
}
# Example usage for your single fits
df1 <- make_df(mu1, 1, "0.0369pM")
df2 <- make_df(mu2, 1, "0.369pM")
df3 <- make_df(mu3, 1, "3.69pM")
df4 <- make_df(mu4, 1, "36pM")
df5 <- make_df(mu5, 1, "369pM")
# Combine all results
all_results <- bind_rows(df1, df2, df3, df4, df5)
# Check
all_results3.5.3 ANOVA and post-hoc multiple comparisons of mumax among varying media treatments:
## Df Sum Sq Mean Sq F value Pr(>F)
## Experiment 4 0.01503 0.003757 1.85 0.196
## Residuals 10 0.02031 0.002031## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = mumax ~ Experiment, data = all_results)
##
## $Experiment
## diff lwr upr p adj
## 0.369pM-0.0369pM 0.079837439 -0.04125361 0.2009285 0.2651098
## 3.69pM-0.0369pM 0.063237082 -0.05785396 0.1843281 0.4652490
## 369pM-0.0369pM 0.089887949 -0.03120310 0.2109790 0.1807721
## 36pM-0.0369pM 0.070735492 -0.05035555 0.1918265 0.3656496
## 3.69pM-0.369pM -0.016600357 -0.13769140 0.1044907 0.9901101
## 369pM-0.369pM 0.010050511 -0.11104054 0.1311416 0.9985578
## 36pM-0.369pM -0.009101947 -0.13019299 0.1119891 0.9990215
## 369pM-3.69pM 0.026650867 -0.09444018 0.1477419 0.9459297
## 36pM-3.69pM 0.007498410 -0.11359264 0.1285895 0.9995435
## 36pM-369pM -0.019152457 -0.14024350 0.1019386 0.98319613.5.4 T-test between 0.0369pM and all other concentrations:
all_results$Group <- ifelse(all_results$Experiment == "0.0369pM",
"A",
"B")
all_results$Group <- factor(all_results$Group)
t.test(mumax~ Group, data = all_results)##
## Welch Two Sample t-test
##
## data: mumax by Group
## t = -1.6943, df = 2.1577, p-value = 0.2231
## alternative hypothesis: true difference in means between group A and group B is not equal to 0
## 95 percent confidence interval:
## -0.2558391 0.1039901
## sample estimates:
## mean in group A mean in group B
## 0.1605090 0.23643353.5.5 Plot mumax with all easy linear data:
# Define the order of treatments from least to greatest
levels_order <- c("0.0369pM", "0.369pM", "3.69pM", "36pM", "369pM")
# Convert Treatment column to a factor with this order
all_results$Experiment <- factor(all_results$Experiment, levels = levels_order)
# calculate standard deviation and mean
b12_summary <- all_results %>%
group_by(Experiment) %>%
summarise(
mean_mumax = mean(mumax, na.rm = TRUE),
se_mumax = sd(mumax, na.rm = TRUE)/sqrt(n()),
.groups = "drop"
)
mu_b12 <- ggplot(all_results, aes(x = Experiment, y = mumax, color = Experiment)) +
# Raw replicate points (jittered)
geom_point(size = 3, shape = 16,
position = position_jitter(width = 0.1), alpha = 1) +
# Mean per Experiment+Round (open circle)
stat_summary(fun = mean, geom = "point", shape = 1, size = 5, stroke = 1.2) +
# Error bars (standard error)
stat_summary(fun.data = mean_se, geom = "errorbar",
width = 0.2) +
# Color palette — WITH legend removed
scale_color_manual(
values = c(
"0.0369pM" = "lightblue",
"0.369pM" = "#99c2e6",
"3.69pM" = "#66a3d2",
"36pM" = "#3380bf",
"369pM" = "#004080"
),
guide = "none" # ← THIS removes the legend safely
) +
# Safe visible range
coord_cartesian(ylim = c(0, 0.4)) +
# Labels + theme
labs(
x = expression(B[12]~Concentration~"(pM)"),
y = expression(paste(mu, " (RFU" %*% "day"^-1, ")"))
) +
theme_minimal() +
theme(
legend.position = "none",
panel.grid = element_blank(),
panel.border = element_rect(fill = NA, color = "black"),
axis.title = element_text(size = 20),
axis.text = element_text(size = 20, color = "black"),
axis.text.x = element_text(size = 20, color = "black", angle = 45, hjust = 1),
strip.text = element_text(size = 20, color = "black"),
axis.ticks = element_line(color = "black", linewidth = 0.5),
axis.ticks.length = unit(4, "pt")
)
mu_b12
3.6 Figure 4
# Ensure BOTH subplots have no legend inside patchwork
max_bio_nl <- max_bio_plot + theme(legend.position = "none")
mu_nl <- mu_b12 + theme(legend.position = "none")
# Build the layout
bo <-
(
growth /
(max_bio_nl | mu_nl)
) +
plot_layout(
widths = c(4, 2), # your original width ratio
guides = "collect" # ← THIS is what actually gathers one legend
) &
theme(
legend.position = "bottom",
legend.box = "horizontal",
panel.spacing = unit(0.4, "lines"),
plot.margin = margin(15, 15, 15, 15),
axis.title.y = element_text(margin = margin(r = 12)),
axis.title.x = element_text(margin = margin(t = 8)),
legend.margin = margin(t = 4, b = 4),
legend.box.margin = margin(5, 5, 5, 5),
legend.ticks = element_blank(),
axis.text.x = element_text(size = 16)
)
bo
Fig. 4 Growth responses of P. bahamense when grown in L1/4 media with a 10% dilution series of B12 concentrations. A In vivo Chlorophyll a fluorescence (Chl-a) was used to track P. bahamense growth in L1/4 media with five concentrations of B12: 0.0369 pM, 0.369 pM, 36 pM, and 369 pM. B Maximum Chl-a values in each replicate by media formulation. Means are plotted as black open circles, and error bars represent one standard deviation of the mean. One-way ANOVA results indicate media had a significant effect on maximum Chl-a (p = 0.0000, F = 31.27 ). Post-hoc multiple comparisons using the Tukey HSD test indicate that the maximum Chl-a achieved in replicates grown in media with the lowest concentration of B12 (0.0369 pM) was significantly lower than in media with 3.69, 36, and 369 pM B12 (p-adj = 0.025, 0.0000, 0.0001, respectively). There were no significant differences in maximum Chl-a achieved between successive B12 concentrations except between 3.69 and 36 pM (p-adj = 0.002). Results from key comparisons are shown on the plot with the following significance codes: p-adj > 0.05; ‘n.s.’, p-adj < 0.05; ’’, and p-adj < 0.001; ’’. C** Specific growth rates (µ; RFU day–1) for each replicate by media formulation. All data are plotted with color representing the media treatment. Means are plotted as black open circles, and error bars represent one standard deviation of the mean. ANOVA results indicate media did not have a significant effect on growth rate (p = 0.196, F = 1.85). However, the growth rates for the replicates grown in media with 0.0369 pM B12 were poor fits (p = 0.1813, R2 < 0.85). The poor fits could have led to overestimated growth rate, obscuring the depressed growth rate in 0.0369 pM B12 and skewing the results of statistical tests.
4 Evaluation of bacterial community potential to supplement B12 and support P. bahamense blooms
Low-diveristy P. bahamense culture was co-cultured with bacteria recruited from OTB (Old Tampa Bay) water, Indian River Lagoon (IRL) culture, and IRL surface seawater and grown in replete media and media missing B12
4.1 Import data:
OTB water:
OTB <- as.data.frame(read_csv("OTB_1006.csv"))
OTB$Time <- as.numeric(sub('.', '', OTB$Time))
OTB$Date <- ymd(OTB$Date)
OTB$Time <- OTB$Time * 48
OTB %<>%
filter(Treatment %in% c("-Cobalamin", "Replete")) %>%
mutate(Treatment = factor(Treatment, levels = c("-Cobalamin", "Replete")))
OTB$Time_in_Days <- OTB$Time / 24IRL water:
IRLrec<-as.data.frame(read_csv("IRL_1006.csv"))
IRLrec$Time <- as.numeric(sub('.', '', IRLrec$Time))
IRLrec$Date <- ymd(IRLrec$Date)
IRLrec$Time <- IRLrec$Time * 48
IRLrec$Time_in_Days <- IRLrec$Time / 24 # Convert Time from hours to daysIRL culture:
4.2 Combine all datasets into one with a new ‘Dataset’ column:
all_experiments <- bind_rows(
fluor %>% mutate(Dataset = "Low Diversity OTB"),
IRLcul %>% mutate(Dataset = "IRL culture microbiome"),
OTB %>% mutate(Dataset = "OTB water microbiome"),
IRLrec %>% mutate(Dataset = "IRL water microbiome")
) %>%mutate(
Treatment = recode(Treatment,
"-Cobalamin" = "\u2013B12",
"-Biotin" = "\u2013Biotin",
"-Thiamine" = "\u2013Thiamine"),
Treatment = factor(Treatment, levels = c("\u2013B12", "Replete"))
) %>%
filter(Round %in% c(1, 2), Treatment %in% c("\u2013B12", "Replete"))4.3 Visulaize growth of P. bahamense with recruited bacteria
# Filter out "Low Diversity OTB"
all_experiments_filtered <- all_experiments %>%
filter(Dataset != "Low Diversity OTB") %>%
mutate(
Treatment = recode(Treatment,
"–B12" = "—B12", # convert en dash to em dash
"Replete" = "Replete"),
Treatment = factor(Treatment, levels = c("—B12", "Replete")),
Dataset = factor(Dataset, levels = c(
"IRL culture microbiome",
"OTB water microbiome",
"IRL water microbiome"
))
)
growth <- ggplot(all_experiments_filtered, aes(x = Time_in_Days, y = `Chlorophyll-A (RFU)`, color = Treatment)) +
geom_point(size = 3) +
geom_smooth(se = FALSE) +
scale_color_manual(values = c("—B12" = "#102E50", "Replete" = "#F5C45E")) +
facet_grid(Round ~ Dataset) +
labs(
x = "Time (days)",
y = expression(paste("Chlorophyll-", italic("a"), " fluorescence (RFU)"))
) +
scale_y_continuous(labels = label_comma()) +
theme_minimal() +
theme(
panel.border = element_rect(fill = NA, color = "black"),
panel.grid = element_blank(),
axis.text.y = element_text(size = 16, color = "black"),
axis.text.x = element_text(size = 12, color = "black"),
axis.title.y = element_text(size = 20, margin = margin(t = 10)),
axis.title.x = element_text(size = 20, margin = margin(t = 10)),
strip.text.x = element_text(size = 20),
strip.text = element_text(size = 20, color = "black"),
legend.title = element_text(size = 16, color = "black"),
legend.text = element_text(size = 16, color = "black"),
legend.position = "bottom",
axis.ticks = element_line(color = "black", linewidth = 0.5),
axis.ticks.length = unit(4, "pt"),
panel.spacing = unit(.1, "cm") ) +
coord_cartesian(ylim = c(0, 5000))
plot(growth)
Create new variable for experimemt and combine experimental data into dataframe:
4.4 Calculate maximum RFU values for each round and medium treatment
max_bio <- all_exp %>%
group_by(Treatment, Bio_Rep,Experiment, Round) %>%
summarise(Max_Chl = max(`Chlorophyll-A (RFU)`, na.rm = TRUE), .groups = "drop") %>%
arrange(Treatment, Bio_Rep)
print(max_bio)## # A tibble: 44 × 5
## Treatment Bio_Rep Experiment Round Max_Chl
## <chr> <dbl> <chr> <dbl> <dbl>
## 1 -Cobalamin 1 IRLcul 1 103.
## 2 -Cobalamin 1 IRLcul 2 136.
## 3 -Cobalamin 1 IRLwat 1 893.
## 4 -Cobalamin 1 IRLwat 2 904.
## 5 -Cobalamin 1 OTB 1 341.
## 6 -Cobalamin 1 OTB 2 87.8
## 7 -Cobalamin 2 IRLcul 1 147.
## 8 -Cobalamin 2 IRLcul 2 173.
## 9 -Cobalamin 2 IRLwat 1 674.
## 10 -Cobalamin 2 IRLwat 2 1126.
## # ℹ 34 more rows4.4.1 Plot maximum biomass
# Define experiment order
experiment_order <- c("fluor", "IRLcul", "OTB", "IRLwat")
# Filter and prepare data for -Cobalamin only
plot_data <- max_bio %>%
filter(Treatment == "-Cobalamin") %>%
mutate(Experiment = factor(Experiment, levels = experiment_order))
#
# plot_summary <- summary_stats %>%
# filter(Treatment == "-Cobalamin") %>%
# mutate(Experiment = factor(Experiment, levels = experiment_order))
# Build plot
max_bio_plot <- ggplot(plot_data, aes(x = Experiment, y = Max_Chl, color = Treatment)) +
# Raw replicate points (solid)
geom_point(size = 5, shape = 16, alpha = 1,
position = position_jitter(width = 0.15)) +
stat_summary(
fun = mean,
geom = "point",
shape = 1,
size = 5,
stroke = 1.2
) +
# Error bars: mean ± SD (NOT SE)
stat_summary(
fun.data = mean_sdl,
fun.args = list(mult = 1), # 1 SD above/below mean
geom = "errorbar",
width = 0.2
) +
# Legend with open circle
scale_color_manual(
values = c("-Cobalamin" = "#102E50"),
labels = c("-Cobalamin" = expression("—B"[12])),
name = "Treatment",
guide = guide_legend(
override.aes = list(
shape = 21,
fill = "white", # open circle in legend
stroke = 1.2,
size = 5
)
)
) +
# Axis labels
labs(
x = "Source",
y = expression(paste("Maximum RFU"))
) +
# X-axis labels
scale_x_discrete(labels = c(
fluor = "Low Diversity",
IRLcul = "IRL Culture",
OTB = "OTB Seawater",
IRLwat = "IRL Seawater"
)) +
# Y-axis formatting
scale_y_continuous(labels = label_comma()) +
# # Facet by Round
# facet_grid(~Round) +
# Theme
theme_minimal() +
theme(
panel.border = element_rect(fill = NA, color = "black"),
panel.grid = element_blank(),
axis.text.y = element_text(size = 20, color = "black"),
axis.text.x = element_text(angle = 45, hjust = 1, size = 14, color = "black"),
axis.title.y = element_text(size = 20),
axis.title.x = element_text(size = 20, margin = margin(t = 10)),
strip.text.x = element_text(size = 20),
strip.text = element_text(size = 20, color = "black"),
legend.title = element_text(size = 16, color = "black"),
legend.text = element_text(size = 16, color = "black"),
legend.position = "bottom",
legend.key.size = unit(1.5, "cm"),
axis.ticks = element_line(color = "black", linewidth = 0.5),
axis.ticks.length = unit(4, "pt")
) +
# Set visible y-axis range safely
coord_cartesian(ylim = c(50, 1500))
# Print plot
max_bio_plot
4.4.2 Run an ANOVA followed by a Tukey HSD test for the experiment factor:
max_B12 <- all_exp %>%
group_by(Experiment ,Treatment, Round, Bio_Rep) %>%
summarise(Max_Chl = max(`Chlorophyll-A (RFU)`, na.rm = TRUE)) %>%
ungroup() %>% filter(Treatment %in% c("-Cobalamin"))
aov_biomass <- aov(Max_Chl ~ Experiment, data =max_B12)
summary(aov_biomass)## Df Sum Sq Mean Sq F value Pr(>F)
## Experiment 3 2481763 827254 38.95 4.45e-08 ***
## Residuals 18 382327 21240
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Tukey multiple comparisons of means
## 95% family-wise confidence level
##
## Fit: aov(formula = Max_Chl ~ Experiment, data = max_B12)
##
## $Experiment
## diff lwr upr p adj
## IRLcul-fluor -148.55167 -414.4357 117.3324 0.4145013
## IRLwat-fluor 679.19000 413.3060 945.0740 0.0000057
## OTB-fluor -23.03833 -288.9224 242.8457 0.9946502
## IRLwat-IRLcul 827.74167 589.9278 1065.5556 0.0000001
## OTB-IRLcul 125.51333 -112.3006 363.3272 0.4626003
## OTB-IRLwat -702.22833 -940.0422 -464.4144 0.00000074.5 Specific growth rates using all_easylinear
4.5.1 Remove outliers from IRL culture data:
outliers are considered values the drop below T0 RFU values
# Compute T0 baseline per replicate, treatment, and round
baseline <- aggregate(
`Chlorophyll-A (RFU)` ~ Bio_Rep + Treatment + Round,
data = subset(IRLcul, Time == 0),
FUN = mean,
na.rm = TRUE
)
names(baseline)[names(baseline) == "Chlorophyll-A (RFU)"] <- "Chl_T0"
# Merge baseline back into the full dataset
IRLcul_merged <- merge(IRLcul, baseline, by = c("Bio_Rep", "Treatment", "Round"), all.x = TRUE)
# Keep all T0 points, and remove any post-T0 points where replicate < its T0
IRLcul_clean <- subset(
IRLcul_merged,
Time == 0 | `Chlorophyll-A (RFU)` >= Chl_T0
)4.5.2 Calculate specific growth rates for IRL culture:
IRL culture trial 1 Replete:
IRLcul_mumax <- IRLcul_clean %>%
filter(Treatment %in% c("Replete"),
Round %in% c(1),
Time >= 0) %>% # <-- exclude T0
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
IRLcul1 <- all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, IRLcul_mumax, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(IRLcul1 , log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(IRLcul1)## $`1:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.03861 0.16034 -0.15112 0.16012 -0.18606 -0.07377 0.12911
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.7592193 0.3438351 5.116 0.00372 **
## x 0.0075785 0.0006407 11.828 7.6e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1627 on 5 degrees of freedom
## Multiple R-squared: 0.9655, Adjusted R-squared: 0.9586
## F-statistic: 139.9 on 1 and 5 DF, p-value: 7.603e-05
##
##
## $`1:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.05977 0.09664 0.02457 0.12515 -0.33697 0.07566 0.07473
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.793494 0.469100 1.692 0.151524
## x 0.007143 0.000691 10.336 0.000146 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1755 on 5 degrees of freedom
## Multiple R-squared: 0.9553, Adjusted R-squared: 0.9463
## F-statistic: 106.8 on 1 and 5 DF, p-value: 0.0001459
##
##
## $`1:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.03760 -0.17761 0.53725 -0.39227 0.01679 0.10783 -0.05440
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.488243 0.895579 1.662 0.15745
## x 0.005008 0.001233 4.062 0.00971 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3132 on 5 degrees of freedom
## Multiple R-squared: 0.7674, Adjusted R-squared: 0.7209
## F-statistic: 16.5 on 1 and 5 DF, p-value: 0.009713## y0 y0_lm mumax lag
## 1:Replete:1 74.22 5.807901 0.007578469 336.1912
## 1:Replete:2 88.61 2.211109 0.007142610 516.7230
## 1:Replete:3 71.52 4.429309 0.005007900 555.4690## 1:Replete:1.r2 1:Replete:2.r2 1:Replete:3.r2
## 0.9654963 0.9552895 0.7674161IRL culture trial 2 Replete:
IRLcul_mumax2 <- IRLcul_clean %>%
filter(Treatment %in% c("Replete"),
Round %in% c(2),
Time >= 0) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
IRLcul2 <- all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, IRLcul_mumax2, h =7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(IRLcul2 , log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(IRLcul2)## $`2:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.23499 0.44031 -0.16656 -0.01381 -0.09558 0.10722 -0.03659
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.7712453 0.4712913 5.880 0.002020 **
## x 0.0069299 0.0009628 7.198 0.000806 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2445 on 5 degrees of freedom
## Multiple R-squared: 0.912, Adjusted R-squared: 0.8944
## F-statistic: 51.81 on 1 and 5 DF, p-value: 0.0008063
##
##
## $`2:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.15234 -0.55703 0.04284 0.55888 0.12964 -0.23617 -0.09050
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.626180 0.806476 2.016 0.09983 .
## x 0.008378 0.001503 5.575 0.00256 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3817 on 5 degrees of freedom
## Multiple R-squared: 0.8614, Adjusted R-squared: 0.8337
## F-statistic: 31.08 on 1 and 5 DF, p-value: 0.002558
##
##
## $`2:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.31177 -0.51835 -0.13281 0.31474 0.18935 -0.07054 -0.09416
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.906698 0.628814 4.623 0.00572 **
## x 0.006738 0.001285 5.245 0.00334 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3263 on 5 degrees of freedom
## Multiple R-squared: 0.8462, Adjusted R-squared: 0.8155
## F-statistic: 27.51 on 1 and 5 DF, p-value: 0.003339## y0 y0_lm mumax lag
## 2:Replete:1 73.64 15.978520 0.006929859 220.4869
## 2:Replete:2 81.69 5.084417 0.008378133 331.4284
## 2:Replete:3 71.42 18.296276 0.006738242 202.1121## 2:Replete:1.r2 2:Replete:2.r2 2:Replete:3.r2
## 0.9119822 0.8614255 0.84622324.5.3 Remove outliers from OTB water data:
# Compute T0 baseline per replicate, treatment, and round
baseline <- aggregate(
`Chlorophyll-A (RFU)` ~ Bio_Rep + Treatment + Round,
data = subset(OTB, Time == 0),
FUN = mean,
na.rm = TRUE
)
names(baseline)[names(baseline) == "Chlorophyll-A (RFU)"] <- "Chl_T0"
# Merge baseline back into the full dataset
OTBwat_merged <- merge(OTB, baseline, by = c("Bio_Rep", "Treatment", "Round"), all.x = TRUE)
# Keep all T0 points, and remove any post-T0 points where replicate < its T0
OTBwat_clean <- subset(
OTBwat_merged,
Time == 0 | `Chlorophyll-A (RFU)` >= Chl_T0
)4.5.4 Calculate specific growth rates for OTB water:
OTB water trial 1 Replete:
OTB_mumax1 <- OTBwat_clean %>%
filter(Treatment %in% c("Replete"),
Round %in% c(1),
Time >= 0) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
OTBwat1 <- all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, OTB_mumax1, h = 7, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(OTBwat1 , log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(OTBwat1)## $`1:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## -0.15859 0.22093 -0.08966 -0.03358 0.16311 -0.01998 -0.08224
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.332309 0.210366 20.59 5.00e-06 ***
## x 0.007191 0.000602 11.95 7.25e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1529 on 5 degrees of freedom
## Multiple R-squared: 0.9661, Adjusted R-squared: 0.9594
## F-statistic: 142.7 on 1 and 5 DF, p-value: 7.251e-05
##
##
## $`1:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.33247 -0.31312 -0.07991 -0.10915 0.06674 0.08438 0.01858
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.6685656 0.3405662 10.772 0.000120 ***
## x 0.0084842 0.0008604 9.861 0.000183 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2185 on 5 degrees of freedom
## Multiple R-squared: 0.9511, Adjusted R-squared: 0.9413
## F-statistic: 97.23 on 1 and 5 DF, p-value: 0.0001828
##
##
## $`1:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6 7
## 0.05109 0.05233 -0.25963 0.01366 0.20820 0.01297 -0.07862
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.3310777 0.2440626 17.75 1.04e-05 ***
## x 0.0068159 0.0006166 11.05 0.000106 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1566 on 5 degrees of freedom
## Multiple R-squared: 0.9607, Adjusted R-squared: 0.9528
## F-statistic: 122.2 on 1 and 5 DF, p-value: 0.0001055## y0 y0_lm mumax lag
## 1:Replete:1 211.13 76.11986 0.007190613 141.8745
## 1:Replete:2 101.61 39.19564 0.008484247 112.2759
## 1:Replete:3 328.76 76.02618 0.006815913 214.8282## 1:Replete:1.r2 1:Replete:2.r2 1:Replete:3.r2
## 0.9661413 0.9510923 0.9606888OTB water trial 2 Replete:
OTB_mumax2 <- OTBwat_clean %>%
filter(Treatment %in% c("Replete"),
Round %in% c(2),
Time >= 0) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
OTBwat2 <- all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, OTB_mumax2, h = 5, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(OTBwat2 , log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(OTBwat2)## $`2:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.04205 -0.14349 0.12612 0.01002 -0.03470
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.0096618 0.0889572 56.315 1.23e-05 ***
## x 0.0041960 0.0007566 5.546 0.0116 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1148 on 3 degrees of freedom
## Multiple R-squared: 0.9111, Adjusted R-squared: 0.8815
## F-statistic: 30.76 on 1 and 3 DF, p-value: 0.01156
##
##
## $`2:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.12058 0.06702 -0.35136 0.01932 0.14443
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.724234 0.180952 26.108 0.000123 ***
## x 0.007660 0.001539 4.977 0.015586 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2336 on 3 degrees of freedom
## Multiple R-squared: 0.892, Adjusted R-squared: 0.856
## F-statistic: 24.77 on 1 and 3 DF, p-value: 0.01559
##
##
## $`2:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.1065 0.1344 -0.4463 0.0634 0.1420
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.409273 0.224859 19.609 0.00029 ***
## x 0.007442 0.001912 3.891 0.03009 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2903 on 3 degrees of freedom
## Multiple R-squared: 0.8347, Adjusted R-squared: 0.7795
## F-statistic: 15.14 on 1 and 3 DF, p-value: 0.03009## y0 y0_lm mumax lag
## 2:Replete:1 156.29 149.85405 0.004195997 10.02180
## 2:Replete:2 127.08 112.64419 0.007660107 15.74165
## 2:Replete:3 91.45 82.20965 0.007442250 14.31284## 2:Replete:1.r2 2:Replete:2.r2 2:Replete:3.r2
## 0.9111292 0.8919806 0.83465014.5.5 Remove outliers from IRL water data:
#Compute T0 baseline per replicate, treatment, and round
baseline <- aggregate(
`Chlorophyll-A (RFU)` ~ Bio_Rep + Treatment + Round,
data = subset(IRLrec, Time == 0),
FUN = mean,
na.rm = TRUE
)
names(baseline)[names(baseline) == "Chlorophyll-A (RFU)"] <- "Chl_T0"
# Merge baseline back into the full dataset
IRLwat_merged <- merge(IRLrec, baseline, by = c("Bio_Rep", "Treatment", "Round"), all.x = TRUE)
# Keep all T0 points, and remove any post-T0 points where replicate < its T0
IRLwat_clean <- subset(
IRLwat_merged,
Time == 0 | `Chlorophyll-A (RFU)` >= Chl_T0
)4.5.6 Calculate specific growth rates for IRL water:
IRL water trial 1 Replete:
IRL_mumax1 <- IRLwat_clean %>%
filter(Treatment %in% c("Replete"), Round %in% c(1)) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
droplevels()
IRLwat1 <- all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, IRL_mumax1, h = 5, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(IRLwat1 , log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(IRLwat1)## $`1:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.07587 0.09142 -0.13804 -0.22946 0.35195
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.005693 0.268748 18.626 0.000338 ***
## x 0.008304 0.001006 8.258 0.003718 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2644 on 3 degrees of freedom
## Multiple R-squared: 0.9579, Adjusted R-squared: 0.9438
## F-statistic: 68.2 on 1 and 3 DF, p-value: 0.003718
##
##
## $`1:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.01569 -0.18763 0.22688 -0.29711 0.27355
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.6061505 0.2231989 25.12 0.000138 ***
## x 0.0064232 0.0008246 7.79 0.004403 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2887 on 3 degrees of freedom
## Multiple R-squared: 0.9529, Adjusted R-squared: 0.9372
## F-statistic: 60.68 on 1 and 3 DF, p-value: 0.004403
##
##
## $`1:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.32328 0.13706 0.35205 -0.08458 -0.08125
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.230102 0.228469 18.515 0.000344 ***
## x 0.013307 0.001943 6.848 0.006374 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.295 on 3 degrees of freedom
## Multiple R-squared: 0.9399, Adjusted R-squared: 0.9198
## F-statistic: 46.89 on 1 and 3 DF, p-value: 0.006374## y0 y0_lm mumax lag
## 1:Replete:1 108.14 149.26049 0.008304417 -38.806618
## 1:Replete:2 267.86 272.09478 0.006423229 -2.442075
## 1:Replete:3 178.44 68.72424 0.013306580 71.705157## 1:Replete:1.r2 1:Replete:2.r2 1:Replete:3.r2
## 0.9578639 0.9528881 0.9398718IRL water trial 2 Replete:
IRL_mumax2 <- IRLwat_clean %>%
filter(Treatment %in% c("Replete"),
Round %in% c(2),
Time >= 0) %>% # <-- exclude T0
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
IRLwat2 <- all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, IRL_mumax2, h =6, quota = 1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(IRLwat2 , log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(IRLwat2)## $`2:Replete:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6
## 0.27656 -0.39465 -0.12399 0.08819 0.39137 -0.23748
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.126211 0.317395 13.000 0.000202 ***
## x 0.009866 0.001698 5.811 0.004365 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3409 on 4 degrees of freedom
## Multiple R-squared: 0.8941, Adjusted R-squared: 0.8676
## F-statistic: 33.76 on 1 and 4 DF, p-value: 0.004365
##
##
## $`2:Replete:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6
## 0.03290 0.04031 -0.06486 0.15651 -0.44416 0.27930
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.883134 0.200405 24.366 1.68e-05 ***
## x 0.006915 0.001379 5.014 0.00741 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2769 on 4 degrees of freedom
## Multiple R-squared: 0.8628, Adjusted R-squared: 0.8284
## F-statistic: 25.14 on 1 and 4 DF, p-value: 0.007414
##
##
## $`2:Replete:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5 6
## -0.10408 0.24742 0.10827 -0.59427 0.39445 -0.05179
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.516376 0.359111 12.577 0.00023 ***
## x 0.008128 0.001921 4.231 0.01336 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3857 on 4 degrees of freedom
## Multiple R-squared: 0.8174, Adjusted R-squared: 0.7717
## F-statistic: 17.9 on 1 and 4 DF, p-value: 0.01336## y0 y0_lm mumax lag
## 2:Replete:1 102.18 61.94276 0.009865993 50.732380
## 2:Replete:2 136.46 132.04390 0.006914858 4.757449
## 2:Replete:3 105.76 91.50335 0.008128438 17.813607## 2:Replete:1.r2 2:Replete:2.r2 2:Replete:3.r2
## 0.8940791 0.8627528 0.8173779IRL water trial 1 -B12:
IRL_mumax1_b <- IRLwat_clean %>%
filter(Treatment %in% c("-Cobalamin"),
Round %in% c(1),
Time >= 0) %>% # <-- exclude T0
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
IRLwat1_b <- all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, IRL_mumax1_b, h = 4, quota =1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(IRLwat1_b , log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(IRLwat1_b)## $`1:-Cobalamin:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## -0.13839 0.02356 0.36806 -0.25322
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.629046 0.277043 16.709 0.00356 **
## x 0.011942 0.003085 3.871 0.06073 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3311 on 2 degrees of freedom
## Multiple R-squared: 0.8822, Adjusted R-squared: 0.8233
## F-statistic: 14.98 on 1 and 2 DF, p-value: 0.06073
##
##
## $`1:-Cobalamin:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## 0.09071 -0.12902 -0.06650 0.10480
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.1181197 0.1239394 41.295 0.000586 ***
## x 0.0053753 0.0007698 6.982 0.019900 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1419 on 2 degrees of freedom
## Multiple R-squared: 0.9606, Adjusted R-squared: 0.9409
## F-statistic: 48.76 on 1 and 2 DF, p-value: 0.0199
##
##
## $`1:-Cobalamin:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4
## 0.06389 -0.33062 0.46957 -0.20284
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.332327 0.711542 4.683 0.0427 *
## x 0.015666 0.004035 3.883 0.0604 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.433 on 2 degrees of freedom
## Multiple R-squared: 0.8829, Adjusted R-squared: 0.8243
## F-statistic: 15.08 on 1 and 2 DF, p-value: 0.06038## y0 y0_lm mumax lag
## 1:-Cobalamin:1 89.18 102.41631 0.011941636 -11.58880
## 1:-Cobalamin:2 182.88 167.02102 0.005375287 16.87548
## 1:-Cobalamin:3 112.12 28.00342 0.015665946 88.55150## 1:-Cobalamin:1.r2 1:-Cobalamin:2.r2 1:-Cobalamin:3.r2
## 0.8822319 0.9605951 0.8828865IRL water trial 2 -B12:
IRL_mumax2_b <- IRLwat_clean %>%
filter(Treatment %in% c("-Cobalamin"),
Round %in% c(2),
Time >= 0) %>%
rename(Chl = `Chlorophyll-A (RFU)`) %>%
arrange(Time, Bio_Rep)
IRLwat2_b <- all_easylinear(Chl ~ Time|Round+Treatment+Bio_Rep, IRL_mumax2_b, h =5, quota =1)
par(mfrow = c(1, 3))
par(mar = c(2, 2, 2, 1))
plot(IRLwat2_b , log = "y")
# Shows the full regression output including coefficients, standard errors, t-tests, p-values, and model fit statistics
summary(IRLwat2_b)## $`2:-Cobalamin:1`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.08915 -0.35337 0.17809 0.34733 -0.26120
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.371518 0.460476 9.493 0.00248 **
## x 0.008698 0.002261 3.847 0.03101 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3432 on 3 degrees of freedom
## Multiple R-squared: 0.8314, Adjusted R-squared: 0.7752
## F-statistic: 14.8 on 1 and 3 DF, p-value: 0.03101
##
##
## $`2:-Cobalamin:2`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## -0.04172 0.31220 -0.17814 -0.37176 0.27941
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.022854 0.290615 17.284 0.000422 ***
## x 0.007184 0.001842 3.899 0.029934 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3402 on 3 degrees of freedom
## Multiple R-squared: 0.8352, Adjusted R-squared: 0.7803
## F-statistic: 15.2 on 1 and 3 DF, p-value: 0.02993
##
##
## $`2:-Cobalamin:3`
##
## Call:
## lm(formula = y ~ x)
##
## Residuals:
## 1 2 3 4 5
## 0.23747 0.05598 -0.51117 -0.09547 0.31319
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.206086 0.396161 10.617 0.00179 **
## x 0.010498 0.002488 4.219 0.02434 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3777 on 3 degrees of freedom
## Multiple R-squared: 0.8558, Adjusted R-squared: 0.8077
## F-statistic: 17.8 on 1 and 3 DF, p-value: 0.02434## y0 y0_lm mumax lag
## 2:-Cobalamin:1 107.43 79.16373 0.008697853 35.103053
## 2:-Cobalamin:2 145.64 151.84401 0.007183847 -5.806909
## 2:-Cobalamin:3 130.09 67.09340 0.010498480 63.070164## 2:-Cobalamin:1.r2 2:-Cobalamin:2.r2 2:-Cobalamin:3.r2
## 0.8314296 0.8352092 0.85575944.5.7 Extract and combine all specific growth rates:
# Helper function to handle matrix or vector coefficients
make_df <- function(model, round_num, experiment_name) {
coefs <- coef(model)
if (is.matrix(coefs)) {
df <- as.data.frame(coefs) %>%
tibble::rownames_to_column("Group") %>%
mutate(
Round = round_num,
Experiment = experiment_name,
r2 = rsquared(model)
)
} else {
df <- data.frame(
Group = names(coefs),
Estimate = as.numeric(coefs),
Round = round_num,
Experiment = experiment_name,
r2 = rep(rsquared(model), length(coefs))
)
}
return(df)
}
# Create data frames for all models with experiment names
df1 <- make_df(IRLcul1, 1, "IRL Culture")
df2 <- make_df(IRLcul2, 2, "IRL Culture")
df3 <- make_df(OTBwat1, 1, "OTB Water")
df4 <- make_df(OTBwat2, 2, "OTB Water")
df5 <- make_df(IRLwat1, 1, "IRL Water")
df6 <- make_df(IRLwat2, 2, "IRL Water")
df7 <- make_df(IRLwat1_b, 1, "IRL Water B")
df8 <- make_df(IRLwat2_b, 2, "IRL Water B")
# Combine all results
all_results <- bind_rows(df1, df2, df3, df4, df5, df6, df7, df8)
# Optional: reorder columns
all_results <- all_results %>%
select(Experiment, Round, Group, y0, mumax, r2)4.5.8 Prepare IRL seawater data for T-test for both trials in -B12 media:
df5 <- as.data.frame(coef(IRLwat1))
df5$Round <- 1
df5$Experiment <- "IRL Water"
df6 <- as.data.frame(coef(IRLwat2))
df6$Round <- 2
df6$Experiment <- "IRL Water"
df7 <- as.data.frame(coef(IRLwat1_b))
df7$Round <- 1
df7$Experiment <- "IRL Water -B12"
df8 <- as.data.frame(coef(IRLwat2_b))
df8$Round <- 2
df8$Experiment <- "IRL Water -B12"
# Combine
IRLwater <- bind_rows(df5, df6, df7, df8)T-test between specific growthrates both trials for IRL water in -B12:
##
## Two Sample t-test
##
## data: mumax by Experiment
## t = -0.59074, df = 10, p-value = 0.5678
## alternative hypothesis: true difference in means between group IRL Water and group IRL Water -B12 is not equal to 0
## 95 percent confidence interval:
## -0.005105451 0.002965605
## sample estimates:
## mean in group IRL Water mean in group IRL Water -B12
## 0.008823919 0.0098938424.5.9 Plot IRL seawater specific growth rates:
# Extract the columns needed
mumax_results <- all_results %>%
select(Experiment, Round, Group, mumax) %>%
mutate(ExperimentRound = paste(Experiment, Round, sep = "_"))
mumax_results <- mumax_results %>%
mutate(
Treatment = factor(
Experiment,
levels = c("IRL Water B", "IRL Water") # order: B12 first, Replete second
)
)
# Filter IRL Water only
mumax_IRLwater <- mumax_results %>%
filter(grepl("IRL Water", Experiment))
IRL_mu <- ggplot(mumax_IRLwater, aes(x = Treatment, y = mumax, color = Treatment)) +
# Raw replicate points
geom_point(size = 5, shape = 16, position = position_jitter(width = 0.1)) +
# Mean per Experiment+Round (open circle)
stat_summary(fun = mean, geom = "point", shape = 1, size = 5, stroke = 1.2) +
stat_summary(fun.data = mean_se, geom = "errorbar", width = 0.2) +
# facet_grid(~Round) +
scale_color_manual(values = c(
"IRL Water B" = "#102E50",
"IRL Water" = "#F5C45E",
"-Thiamine" = "deeppink4",
"-Biotin" = "#667028"
)) +
scale_x_discrete(labels = c(
"IRL Water B" = expression("\u2013B"[12]),
"IRL Water" = "Replete",
"-Thiamine" = expression("\u2013Thiamine"),
"-Biotin" = expression("\u2013Biotin")
))+
theme_minimal() +
theme(
legend.position = "none",
panel.grid = element_blank(),
panel.border = element_rect(fill = NA, color = "black"),
axis.title = element_text(size = 20),
axis.text = element_text(size = 20, color = "black"),
axis.text.x = element_text( size = 18, color = "black", angle = 45, hjust = 1),
strip.text = element_text( size = 20, color = "black"),
legend.title = element_blank(),
legend.text = element_blank(),
legend.key = element_blank(),
axis.ticks = element_line(color = "black", linewidth = 0.5),
axis.ticks.length = unit(4, "pt")
) +
labs(
x = "Media Treatment",
y = expression(paste("(", mu, ") RFU" * "\u00B7" * "day"^-1, " "))
) +
guides(color = "none", fill = "none", shape = "none") +
# Set visible range safely
coord_cartesian(ylim = c(0, 0.5))
IRL_mu
4.6 Figure 5
# Ensure BOTH subplots have no legend inside patchwork
max_bio__2 <- max_bio_plot + theme(legend.position = "none")
mu_2 <- IRL_mu + theme(legend.position = "none")
bo <-
growth /
(max_bio_plot +
IRL_mu ) +
plot_layout(
guides = "collect",
widths = c(4, 2)
) &
theme(
legend.position = "bottom",
panel.spacing = unit(.1, "lines"),
plot.margin = margin(15, 15, 15, 15),
axis.title.y = element_text(margin = margin(r = 12)),
axis.title.x = element_text(margin = margin(t = 8)),
legend.margin = margin(t = 4, b = 4),
legend.box.margin = margin(5, 5, 5, 5),
axis.text.x = element_text(size = 16)
)
bo
Fig. 5 Growth responses of P. bahamense (FWC1006, OTB) with bacterial communities recruited from FWC1003 (IRL) P. bahamense culture, OTB seawater, and IRL seawater when grown in L1/4 media missing vitamin B12 and replete L1/4 media A In vivo Chlorophyll a fluorescence (Chl-a) was used to track the growth of P. bahamense (FWC1006, OTB) with bacterial communities recruited from three sources in L1/4 media missing B12 (yellow) and replete L1/4 media (blue) for two experimental rounds. Replicates growing in replete media and replicates with bacteria recruited from IRL seawater were serially transferred from experimental round one to round two. The biomass in replicates with bacteria from the FWC1003 P. bahamense culture and from OTB seawater grown in media missing B12 were too low to support serial transfers; the replicates in these treatments were restarted from cultures with the same bacterial communities in replete media for round two. B Maximum Chl-a values in each replicate grown in media missing B12 by bacterial community source. The maximum Chl-a reached in media missing B12 with the initial low-diversity bacterial community in the FWC1006 P. bahamense culture (from figure 3) is included for reference. One-way ANOVA results indicate that bacterial community source had a significant effect on maximum Chl-a when vitamin B12 was excluded from media (p = 0.0001, F = 50.27). Post-hoc multiple comparisons (Tukey HSD) showed that the cultures with bacteria from IRL seawater had significantly higher maximum Chl-a compared to cultures with all other bacterial communities (p-adj = 0.0000 for each comparison). The maximum Chl-a was not significantly different between cultures with the other three bacterial communities. Results from key comparisons are shown on the plot with the following significance codes: p-adj < 0.001; ’*’. C** Specific growth rates (µ; RFU day–1) for each replicate with the bacterial community recruited from IRL seawater by media formulation. Means are plotted as black open circles, and error bars represent one standard deviation of the mean. The growth rates for FWC 1006 P. bahamense with bacteria from IRL seawater were not significantly different in media with or without B12 (two-sample t-test; p = 0.5678, t(10) = –0.5907).
5 Characterizing P. bahamense recruited bacterial communities
- Full-length 16S rRNA was sequenced using Oxford Nanopore MinIon v10.4 flow cells (Oxford Nanopore Technology 2008)
- Sequence data was processed and classified using Kraken v2.1.2 (Lu and Salzberg 2020)
5.2 Convert to long data:
# Identify sample columns
sample_cols <- names(Recruit_16s)[2:(which(names(Recruit_16s) == "total") - 1)]
# Pivot to long data
Recruit_16s_long <- Recruit_16s %>%
pivot_longer(
cols = all_of(sample_cols),
names_to = "sample",
values_to = "count"
)
# Separate samples into new columns
Recruit_16s_long <- Recruit_16s_long %>%
separate(sample, into = c("experiment","treatment","filter_size","bio_rep"),
sep = "_", remove = FALSE)
# One genus in one sample with the total count of that genus in that sample
grouped_data1 <- Recruit_16s_long %>%
mutate(count = as.numeric(count)) %>%
group_by(treatment, experiment, filter_size, bio_rep, genus) %>%
summarise(total_count = sum(count), .groups = "drop"
)5.3 Prepare data as wide matrix:
# Pivot wider
wide_data1 <- grouped_data1 %>%
pivot_wider(names_from = genus, values_from = total_count, values_fill = 0)
# Create rownames and remove metadata columns for distance matrix
rownames(wide_data1) <- paste(wide_data1$treatment, wide_data1$experiment, wide_data1$filter_size, wide_data1$bio_rep, sep = "_")
wide_matrix1 <- wide_data1[, !(names(wide_data1) %in% c("treatment", "experiment", "filter_size", "bio_rep"))]
wide_matrix1 <- as.data.frame(lapply(wide_matrix1, as.numeric))
wide_matrix1[is.na(wide_matrix1)] <- 05.4 Compute Bray–Curtis Distances and Perform PCoA
# Bray-Curtis + PCoA
bray_dist1 <- vegdist(wide_matrix1, method = "bray")
pcoa_result1 <- pcoa(bray_dist1)
# Build plot data
pcoa_data1 <- as.data.frame(pcoa_result1$vectors)
pcoa_data1$treatment <- wide_data1$treatment
pcoa_data1$experiment <- wide_data1$experiment
pcoa_data1$filter_size <- as.factor(wide_data1$filter_size)5.5 Calculate Permanova:
#ensure wide matrix is numeric
wide_matrix1 <- as.data.frame(lapply(wide_matrix1, as.numeric))
wide_matrix1[is.na(wide_matrix1)] <- 0
#Build the Bray–Curtis distance matrix
bray_dist <- vegdist(wide_matrix1, method = "bray")
#Bring metadata back in (from wide_data1)
metadata <- wide_data1[, c("treatment", "experiment", "filter_size", "bio_rep")]
permanova_result <- adonis2(
bray_dist ~ experiment,
data = metadata,
permutations = 999
)
print(permanova_result)## Permutation test for adonis under reduced model
## Permutation: free
## Number of permutations: 999
##
## adonis2(formula = bray_dist ~ experiment, data = metadata, permutations = 999)
## Df SumOfSqs R2 F Pr(>F)
## Model 4 6.2820 0.67542 16.127 0.001 ***
## Residual 31 3.0189 0.32458
## Total 35 9.3009 1.00000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 15.6 Figure 6 PCoA plot
sw_pcoa <- ggplot(
pcoa_data1,
aes(
x = Axis.1,
y = Axis.2,
color = experiment,
shape = treatment,
alpha = filter_size # keep as factor
)
) +
geom_point(size = 5) +
labs(
x = paste0("PCoA1 (", round(pcoa_result1$values$Relative_eig[1] * 100, 2), "%)"),
y = paste0("PCoA2 (", round(pcoa_result1$values$Relative_eig[2] * 100, 2), "%)"),
color = "Experiment",
shape = "Treatment",
alpha = "Filter size"
) +
stat_ellipse(aes(group = experiment)) +
scale_color_manual(values = pp_palette)+
# Map discrete alpha manually
scale_alpha_manual(values = c("02" = 0.4, "10" = 1)) +
theme_minimal() +
theme(
legend.key.size = unit(1.5, "cm"),
legend.title = element_text(size = 16),
legend.text = element_text(size =12),
panel.border = element_rect(color = 'black', fill = NA),
axis.title.x = element_text(size = 16, color = "black"),
axis.title.y = element_text(size =16, color = "black"),
strip.text = element_text(size =12, color = "black"),
axis.text.y = element_text(size = 16, color = "black"),
axis.text.x = element_text(size = 16, color = "black")
) +
geom_hline(yintercept = 0, linetype = "dotted", linewidth = 0.5) +
geom_vline(xintercept = 0, linetype = "dotted", linewidth = 0.5) +
theme(panel.grid = element_blank())
plot(sw_pcoa)
5.7 Prepare data for relative abundance plot:
#Standardize experiment names
grouped_data <- grouped_data1 %>%
mutate(
experiment = case_when(
grepl("1003.1006", experiment, ignore.case = TRUE) ~ "1003.1006",
grepl("OTB", experiment, ignore.case = TRUE) ~ "OTB.1006",
grepl("IRL", experiment, ignore.case = TRUE) ~ "IRL.1006",
TRUE ~ experiment
),
experiment = factor(experiment,
levels = c("1003.1006", "OTB_seawater", "OTB.1006", "IRL_seawater", "IRL.1006"),
labels = c("Microbiome from IRL culture",
"OTB Seawater",
"Microbiome from OTB seawater",
"IRL seawater",
"Microbiome from IRL seawater"))
)
#Pivot to long format
sample_cols <- names(Recruit_16s)[2:(which(names(Recruit_16s) == "total") - 1)]
Recruit_16s_long <- Recruit_16s %>%
pivot_longer(
cols = all_of(sample_cols),
names_to = "sample",
values_to = "count"
) %>%
separate(sample, into = c("experiment","treatment","filter_size","bio_rep"),
sep = "_", remove = FALSE) %>%
mutate(count = as.numeric(count)) # ensure numeric
# ️ Summarize counts per sample & genus
grouped_data <- Recruit_16s_long %>%
group_by(experiment, treatment, filter_size, bio_rep, order) %>%
summarise(total_count = sum(count), .groups = "drop") %>%
group_by(experiment, treatment, filter_size, bio_rep) %>%
mutate(RA = (total_count / sum(total_count)) * 100) %>%
ungroup() %>%
filter(treatment %in% c("noB12", "replete", "seawater"))
# ⃣Collapse low abundance genera (<1%) into "z<1% abund."
grouped_data$order <- as.character(grouped_data$order)
grouped_data$order [grouped_data$RA < 1] <- "z<1% abund."
grouped_data$order <- factor(grouped_data$order ) # back to factor
# Optional: fix experiment & filter labels for plotting
grouped_data <- grouped_data %>%
mutate(
experiment = factor(experiment,
levels = c("X1003.1006","OTB","OTB.1006","IRL","IRL.1006"),
labels = c("Recruited IRL culture",
"OTB Seawater",
"Recruited OTB seawater",
"IRL Seawater",
"Recruited IRL seawater")),
Treatment = recode(treatment, "-B12" = "\u2014B12"),
filter_size = recode(filter_size, "10" = "10.0um", "02" = "0.2um"),
filter_size = factor(filter_size, levels = c("10.0um", "0.2um")), # <-- top first
x_label = paste(Treatment, "Rep", bio_rep, sep = " ")
)
# Define color palette (ensure enough colors for all genera)
unique_genera <- levels(grouped_data$order )
n_colors <- length(unique_genera)
palette_colors <- brewer.pal(min(12, n_colors), "Set3") # adjust if >12 genera
if(n_colors > 12) {
palette_colors <- colorRampPalette(palette_colors)(n_colors)
}
names(palette_colors) <- unique_genera5.8 Figure 7 Relative abundance plot
n_orders <- length(unique(grouped_data$order))
# expand the 5-color palette into n_orders colors
order_palette <- colorRampPalette(pp_palette)(n_orders)
taxa_bar_plot <- ggplot(grouped_data,
aes(x = x_label, y = RA, fill = order)) +
geom_bar(stat = "identity", position = "stack") +
facet_grid(filter_size ~ experiment, scales = "free_x") +
theme_classic() +
theme(
legend.position = "right",
legend.title = element_blank(),
axis.text.x = element_text(size = 18, angle = 45, hjust = 1, color = "black"),
axis.text.y = element_text(size = 18, color = "black"),
axis.title.y = element_text(size = 18),
strip.text = element_text(size = 12),
legend.text = element_text(size = 18),
strip.position = "top"
) +
scale_fill_manual(values = order_palette) +
guides(fill = guide_legend(ncol = 2)) +
ylab("Relative Abundance (%)") +
labs(x = NULL)
taxa_bar_plot